http://jossviz.wordpress.com/2010/06/23/friendwheel-layout-of-a-facebook-network/

The Journal of Social Structure (JoSS) is an electronic journal of the International Network for Social Network Analysis (INSNA).  Here is Bernie’s description of the graph.

This is a “pinwheel” diagram using the author’s Facebook personal network (captured July 15, 2009).

Nodes represent the author’s friends and links represent friendships among them. The author is not shown. Each ‘wing’ radiating outwards is a partition using a greedy community detection algorithm (Wakita and Tsurumi, 2007). Wings are manually labelled. Node ordering within each wing is based on degree. Node color and size is also based on degree. Nodes position is based on a polar coordinate system: each node is on an equal angle of n/360º with a radius being a log-scaled measure of betweenness. Higher values are closer to the center indicating a sort of cross-partition ‘gravity’.

This layout has several notable features:

- The angle of each wing is proportionate to its share of the network. Thus 25 percent of nodes go from 0 to 90º.

- Partitions are distinguished by their position rather than a node’s color or shape.

- The tail indicates the periphery of each partition. A wing with many tail nodes indicates many people who are only tied to other group members.

- Edges crossing the center show between-partition connections. Since nodes are sorted by degree it is easy to see if edges originate from the most highly connected nodes or the entire partition.

Bernie’s chapter on analyzing Facebook networks with NodeXL appears in the book: Analyzing Social Media Networks with NodeXL: Insights from a connected world.

A new paper on visualizing social media has been released on the University of Maryland, Human Computer Interaction Laboratory tech report archive.  Co-authored by Derek Hansen,  myself, and Ben Shneiderman, the paper describes and visualizes the patterns of connections formed when people tweet about events like conferences and news stories.

EventGraphs_2010_HCIL_Tech_Report

http://www.cs.umd.edu/localphp/hcil/tech-reports-search.php?number=2010-13

Hansen, D., Smith, M., Shneiderman, B.
EventGraphs: Charting Collections of Conference Connections
HCIL-2010-13

EventGraphs are social media network diagrams constructed from content selected by its association with time-bounded events, such as conferences. Many conferences now communicate a common “hashtag” or keyword to identify messages related to the event. EventGraphs help make sense of the collections of connections that form when people follow, reply or mention one another and a keyword. This paper defines EventGraphs, characterizes different types, and shows how the social media network analysis add-in NodeXL supports their creation and analysis. The paper also identifies the structural and conversational patterns to look for and highlight in EventGraphs and provides design ideas for their improvement.

Pierre has been a deep student of telecommunications policy regulation in the United States for many years.  He has generated a remarkable network map built from the details of filings to the FCC over more than a decade.  These filings are made by companies when they agree or disagree with a proposed policy.  When two companies file in support (or opposition) to the same policy they create a tie between them.  The collection of these connections creates a complex network of coalitions and factions.

http://www.cmu.edu/joss/content/issues/2010jossviz/5_deVries.htm

“The graph is derived from meta-data associated with documents that are filed electronically whenever an organization interacts with the FCC, in accordance with the Administrative Procedures Act. Whenever a letter, comment or other document is filed, the filer provides information on the parties involved, number of pages, relevant proceedings, date, etc.”

…

“Once the data is cleaned up, an edge list is created in Excel by running another VBA macro. A graph is created from this list with NodeXL, a social network analysis and visualization add-in for Excel 2007. NodeXL’s Fruchterman-Reingold algorithm is used to prepare a preliminary layout; nodes are then moved by hand into visually intelligible positions, respecting the clusters suggested by NodeXL’s implementation of the Wakita-Tsurumi algorithm. Nodes are colored on the basis of eigenvector centrality. The degree of investment that organizations make in lobbying is measured by the total number of filings it made in this proceeding over the period of study, and reflected in the size of the node. This information is obtained by running another VBA macro against the underlying ECFS metadata, and then matching that to the vertices in the graph.”

Read more about this industry network at JoSS.

Fourth International AAAI Conference on Weblogs and Social Media (ICWSM-10)

Michael Kearns Keynote

Experiments: Graph Coloring / Consensus / Voting

Topology of the Network vs. what was the network used for?

Voting experiments – similar to consensus, with a crucial strategic difference.

Introduce a tension between:

-Individual preferences

-Collective unity

-Color choices; challenge comes from competing incentives

Red, blue. People unaware of global network structure

Payoffs: if everyone picks same color w/in 2 minutes, experiment ends, and everyone gets some payoff. But different players have different incentives (e.g. I may get paid p if everyone converges to blue, but 2p if everyone converges to red). If there is no consensus, nobody gets a payoff

Systems point: confuse player perceptions of system so that players don’t lock into a consensus early on

Results: Varied homophily, random vs. PA ties. 27 total experiments

Result 1: ~70% experiments solved

Result 2: a minority of hubs in a PA networks will dictate the preferences of the network (24/27 experiments converged, 100% of converged picked the minority preference)

In general, it seems to help (in terms of decreasing convergence time) to have one part of the population “care more” about their preference

Effects of “Personality” – people will be stubborn and hold out for their color even when it’s clearly in the minority.

Lessons Learned, 2005-2009

1. People are remarkably good over large set of collective tasks and network topologies (over all experiments, efficiency close to 90%)

2. Network structure matters, often on a task-specific basis

3. Problem – exogenously imposing network on subjects

-are “hard” network structures just unlikely to arise in the real world?

–Network formation games

New experiments: biased voting game + network formation

-everybody starts off as a single vertex and can’t see anyone else’s color

-at any point, players can spend money to purchase edges (money deducted from final winnings in game)

-you are shown all your neighbors, plus all other nodes in a grid. For nodes that are not your neighbors you are shown their degree and current distance away from you

Strategic tensions:

1. Should you buy edges or not? Ideally, want neighbors to buy edges for you, but need a MST to coordinate on task

2. Buy edges for information or for influence?

3. Buy early or late?

4. Buy from high degree or low degree people?

Experimental Designs: 63 experiments, no network to begin with. Additionally, ran 36 experiments where a network structure existed at beginning of experiment but edges could still be bought

Early results: Subjects do quite poorly at network formation games relative to any previous experiments! (47% in first set of tasks, 38% in second set of tasks)

–preliminary evidence shows that people are building networks that make it difficult to solve the biased voting problem

***Sentiment and Language Analysis***

ICWSM – A Great Catch Name: Semi-Supervised Recognition of Sarscastic Sentences in Online Product Reviews (Tsur et al.)

Challenge: Many different definitions beyond the basic one

–Context

–World knowledge

How do people cope?

-Temherte slaq (Some Ethiopic Languages): inverse exclam

-Reverse question mark

-#sarcasm

Data:

-Amazon product reviews (~66K)

–Books, Electronics

-Additional study based on ~6 mln tweets

Star Sentiment Baseline (Amazon)

-”Saying or writing the opposite of what you mean”

–Find unhappy reviewers, look for overwhelmingly positive sentiment

SASI: Semi-supervised Algorithm for Sarcasm Identification

-Label sarcasting-tagged sentences. Tags 1-5 (for different levels of sarcasm)

-Extract features from all training sentences

-represent training sentences in feature space, do KNN

Preprocessing: [author],[title],[product],[company]

Pattern-based features:

–High Frequency Words,

–Content Words

–pattern e.g. {[Frequent] [CW]}*

Weights of pattern based features:

-1: exact match

-alpha – extra elements are found between components

-gamma – incomplete match

Punctuation based features: Number of !, CAPITALIZED words/letters

Classification: weighted-kNN

Experiment 1: 5-fold cross validation on training set: F Score up to .827

Experiment 2: Gold Standard evaluation

–Human annotation of classification of new sentences: F Score up to .788

—F Score improves if you use algorithm on Tweets! (to .827)

Widespread Worry and the Stock Market (Gilbert and Karahalios)

Lab experiments in psych & behavioral econ

–Emotions affect our choices at dcision time

–Fear affects our choices, makes us risk-averse

If we estimate worry and fear, can that tell us anything about the stock market?

–Stock market is probably not efficient (e.g. more likely to go up on a sunny day than down)

–Online media have predictive information

Data: 2008 Livejournal: Feb-Jun, Aug-Sep, Nov-Dec

Why LJ? Place where people talk about their daily lives

Training data: the anxiety index

620K mood-annotated LJ posts. Picked “anxious, worried, nervous, fearful.” = 13K

C1 = Boosted decision tree with top 100 stems

C2 = Complement Naive Bayes

Both classifiers have low true positive rates

Re-mapped to low-frequency data: max of both classifier to label trading day t

Market data: SP_t = S&P-500 closing price

Controlled for volume and volatility of stock market

Method: Granger Causality (Autoregressive Approach, F test)

Result: Adding in anxiety index explains more significantly variance than baseline autoregressive model

claim: estimating worry and fear seems to have some information about market direction

Star Quality: Aggregating Reviews to Rank Products and Merchants (McGlohon, Glance, Reiter)

Google product search

The problem: given reviews, aggregated from different sources, how to measure “true quality” of product. What is the gold standard?

Challenges:

-Different sources have different review scales

-Different sources have different rating distributions

-Reviews may be plagiarized or irrelevant (cf. Danescu-Niculescu-Muzyl 2009)

Outline:

-Analyze ratings aggregated from many review sites

-Propose models to determine “true quality”

-Build evaluation framework

Data:

-Product reviews: 8M ratings (560K products, 3.8M products, 230 sources)

Observation 1: People like passing out 5′s, single-review authors disproportionately more so

Observation 2: Authors / Sources have biases

-Ratings for same product differ widely

-Authors are consistent across products (Like everything or hate everything)

-Sites vary (pricegrabber = 4.5 stars, another site = 2.9 stars average)

Observation 3: The rated object matters

Merchant reviews more “binary”

Netflix more “normal”

Observation 4: How much an object is rate matters (rich-get-richer)

Proposed Models:

1. Mean rating for an object (baseline)

2. Median rating for an object

3. Lower bound on normal confidence interval

4. Binomial confidence interval

5. Average percentile of order statistic (“most websites liked it better than other products”)

6. Filtering anonymous reviews, then average

7. Filter prolific authors, then average

8. Rate authors by reliability

Evaluation Method

-No “ground truth” for quality

-Goal: to see how reliably our ranking of “true quality” agrees with user preferences

-Hold out a pair of ratings from the same author, test on the hold-outs

-For every “prolific” author  hold out two pairs of reviews at random for test data

-Then in training data, calculate estimated quality, rank objects accordingly

-Then compare the given ranking with ranking in each pair in test data

-Results: No method significantly outperforms average rating!

Fourth International AAAI Conference on Weblogs and Social Media (ICWSM-10)

We will be liveblogging (when possible) from ICWSM 2010, going on now!

Keynote: Bob Kraut, CMU

kraut keynote
implications for community design-offline theories of socialization helpful, not definitive
-online communities can build in good socialization practice-e.g. WP welcoming committee
Two Types of Commitments to Groups-identity based groups-bond based groups
Added Identity & Bond Features to MovieLens
Introduced Subgroups into MovieLens
Identity features that focus on subgroups
Individual profiles
bond-based design:+11% loginsidentity-based design:+44% logins
interventions based on theory increased commitmen
why stronger effect of identity?-time course: social identity can form instantaneously, bonds take time
new approaches to translate theory to design-ABMs
test identity vs. bond design via abm
incorporate design into abm
what are the consequences of discussion moderation-what type of moderation should be imposed? when?
results: indiv moderation helps logins
-personalized mod improves info and social benefit
-comm level mod improves info benefit only in homogeneous communities

***Influence and Composition in Social Networks***

Measuring Influence of Neighbors in Social Networks (Cosley Huttenlocher Kleinberg, Lan Suri)

Measuring Influence in 2 types of socnet data

-Crawl

–provides state of network at specific time

–often have more than one

-Complete time seties

–often from database dump

–provides timestamped history of all events

-This talk: How do measurements of social influence in these two settings compare?

Defining Influence

p(k) = Pr(individual adopts new behavior | k neighbors have)

Analyzing Influence in Wikipedia

-Wikipedia as a social network

–registered users (500K) represented by odes

–communicate through user talk pages

—Definition: Link between u and v forms at time t if one edited the other’s talk page at time t

-Definition: A community is a set of users who edited an article

-Analyze: Pr(user joins a community | k of his friends have)

Ordinal Time Method for Measuring Influence:

-p_o(k) = # instances (over all C) where u had k neighbors in C and joined/  # instances (over all C) where u had k neighbors in C

-advantage: very easy to interpret

-disadvantage: requires complete time series

Snapshot Time Method for Measuring Influence:

-2 snapshots of the network taken at times t1 and t2

p_s(k) = # instances (over all C) where u had k neighbors in C at t_1 and joined before t2 / #instances (over all C) where u had k neighbors in C at t1 and did not join before t1

-advantages: requires only a snapshot

-disadvantage: coarse-grained, don’t know changes in friends between t1 and t2

-results: p_o(k) behaves very differently from p_s(k) on Wikipedia

How are p_o(k) and p_s(k) different?

-accumulation effect in p_o(k) where community joining events can contribute to p_o(1)…p_o(k)

Approximating Ordinal Time from Snapshots

-At each snapshot know:

–which users joined which communities

–how many friends each user had in each community

-Don’t know exact number of friends user had when joined

-Assume a constant rate of getting friends

Ethnicity in Social Networks (Facebook Data Team)

-Questions about how ethnicities engage online are prevalent

-Our goal is to better understand ethnicity on social networks

-Issues

–How do we estimate ethnic distributions on social networks?

–Many services (e.g. Facebook) do not ask their users about their ethnicity

–Our approach infers ethnicities from surnames (based on previous approaches)

Topic modeling approach

-words are names, topics are ethnicities

-Generative process: Draw ethnic bareakdown of aggregate population w/Dirichlet distribution

–For each person draw ethnicity of individiual z_n ~ Multinomial

–Draw surname of individual based on ethnicity ~ Multinomial conditioned on ethnicity

Because there is no FB ground truth, we test our method by scraping 10k users from Myspace

Results: Model does a lot better than naive guessing based on census, internet usage statistics

Also tested on FB data (no ground truth but can have qual results):

1. In 2006, AsAm and White overrepresented on FB, by 2008 ethnicities much closer to mark

2. Drew heatmaps of US, showed that Whites concentrated in the northeast, Asians on the coasts, etc.

3. Correlated with politics: AsAm more likely to be liberal, Whites more likely to be conservative/libertarian

Homophily:

-Strong racial homophily (strongest in Hispanics and Blacks, weakest in Whites)

Homophily by ordinal friendships: racial homophily is strongest for the first few friendships a person makes

Friend you communicate with the most is most likely to be homophilous friend

Who makes friends in social media and why? Rich get Richer vs Seek and Ye Shall Find (Zeynep)

Importance of friendship and social networks vs. debate about social isolation

Architecture of our Commons

-Sociality does not happen in a vacuum

-Mediated Relationships (through digital means)

–Are these relationships: Superficial, weak, fake? Supplementary to offline sociality? As real or more?

–Why does this debate refuse to die?

-Assumptions:

–Internet communication: Anonymous, text-based, fleeting

–Lack of visual cues

but today’s Internet:

–social media: non-transitory interactions, lots of visual cues

Two dominant theories: Rich get Richer vs. Seek and Ye Shall Find

Data: Sample:

-College students

-617 respondents

-multiple classrooms

-very diverse school

-diverse majors

Results: feeling about whether online friendships are possible vs. not possible – split down the middle

Qualitative Component: why do respondents feel online friendships possible / not possible

*Not Possible reasons:

-trust

-face-to-face

-body language and mannerisms

-shared emotions and experiences

*Possible:

-deeper connections – easier (“it may even be easier online as it is all dialogue and no physical characteristics involved”)

-deeper connections – judgment

-bonding is possible

-conversation as key

-experience

***Diffusion and Dynamics in Networks***

Social Dynamics of Activity in a Virtual World (Bakshy et al.)

Second Life paper

Second Life: persistent, immersive virtual world

-User driven objects, economy, society

RQs: 1. How much economic activity occurs in virtual world?

2. Role of groups?

3. How much virtual interaction?

Data:

a) 65 mln user-user transactions (virtual goods)

b) Buddy graph – 4.2 million users, 43 million relationships, focus on strong edges: reciprocal and permissive (can see each other’s online status)

c) chat between 14 million pairs of users

d) 520k groups, 23 million user/group memberships

Economic Activity in Second Life

-Same “sectors” as realworld: retail, real estate, entertainment

-29 million free transactions, 36 million paid, power-law distribution of exchange amounts

Seller in detail: shows that interaction does not necessarily correspond to exchange

Analysis hints at Seller roles? Profiles?

Social Adoption

-show how friends of friends of initial buyers go on to buy as well. Is it influence or homophily? Not sure.

Role of Social ties

-39% of free transactions (but only 7% of paid) were between friends

-40% of users that chat exchange free items, 12% of users that chat engage in paid transactions

Free transactions may be more representative of social activity than paid transasctions

Role of Groups

-Groups indicate aspect of user interest

-Co-grouped transactions

-Sellers more likely to be connected to buyers through co-group than through friendship

Long discussion of why groups are proxy for connections

Explaining Seller Success

Traditional success measures: revenue, repeat business. Regression analysis

Key predictors:

For Revenue:

-Amount made by friends (homophily? business partnerships?)

-Make connections

-Less chatting with customers

-The younger crowd (newer to 2nd Life)

For Repeat Business:

-Interaction!

–Sharing a group, chatting

-More established crowd

-More diverse buyer base

Your Brain on Facebook (Fisher and Counts)

EEG: social vs. traditional media

Why do this? Possible input/feedback to social interaction systems. Short term: Inform design through better understanding of automatic info processing

EEG very good at detecting semantic mismatch

RQs: is Myspace associated with frivolity? Is connecting with friends on Facebook connected with feelings of intimacy?

Media and concepts:

-Media: TV, Books, Social, News

-Concepts: addictive, story, interesting, frivolous, personal, useful

-Showed both front page of FB and personal FB page (asked study participants to briefly friend study on Facebook)

Measurement and Method:

1. Timed, binary decision (low conscious reasoning)

2. Lijkert scale untimed survey (high conscious reasoning)

-Computer-based task: 24 combinations x 22 trials = 528 trials

-16 participants

Results (survey): FB more addictive, but less useful than news. Tells less of a story than books or TV. More frivolous than books, news. More personal than all other forms of media.

Results (decision taks): FB very addictive, tells less of a story than books or TV, a lot more personal than other forms of media. Overall, quite similar questionnaire responses.

Viz: heatmap of head. Time chart. In time chart, look for potential drop ~400 ms after stimulus. The more similar the media to the concept, the bigger the drop.

Results (eeg): FB equal to other media in terms of addictive, interesting, useful, frivolous. Tells less of a story than other media. Is less personal than other media.

Implications, suggestions, limitations

Media parity: FB as interesting, useful, addictive and not frivolous as other media

Personalization and self-identification: Purposeful connection building; easy switching to close friends, family; identity vs. bond attachment

Form of media found most personal by eeg? Books!

Telling stories – status updates on storylines?

Making it tangible: What’s the ratty, marked-up favorite book equivalent in social media? Photo album equivalent?

Limitations: only one form of social networking / media, limited subject population, can we believe EEG results?

Opportunities: compare online media, other subject populations, corroborate with other physiology, expand into real-time capture of physiology

Social Causality and Analysis of Interpersonal Relationships in Online Blogs and Forums (Roxana Girju)

Social causality: causal reasoning used by intelligent agents in a social environment

Modeling social causality – can guide conversation strategies, facilitate modeling and understanding of social emotions, bring new insights

Our focus: social causality as capture through analysis of interpersonal relations in social media

-Pervasive set of english reciprocal textual contexts encoding interpersonal relationships

-data: 11K reciprocal relationship contexts coded

Reciprocity in Language: “The Golden Rule”

Modeling relationships

Properties of interpersonal verbs and reciprocal instances: Symmetry, Affective Value (4-state HMM), Intentionality of Actions

-Intentionality and affective values of interpersonal verbs highly correlated with blame and responsibility

Analyzing Social Interactions:

-Overall 54.34% of dataset was encoded by ambiguous symmetric patterns

-top frequent verb pairs: need-need, love-love

-followed by: hate-hate, miss-miss…

-In general we love people who love/understand/care/need us

-Gender analysis: men initiate more often than females, retaliate more often than women, are more violent and aggressive (whereas women are more forgiving), but this depends on class of verb

-Men and women generally mutually respectful, it is only when respect is broken that responses may differ (e.g. women: cheat -> hate, despise, sue. men: cheat -> dump, divorce)

-Intentionality of actions: intentionality much more often perceived as intentional in bad-bad exchanges than in good-good exchanges

-Reciprocity Chains: Dyadic (formed between 2 people of the form A v B -> B v A -> A v B). Very useful in micro-levle social interaction analysis. General (between multiple people)

In generic chains: retaliation with increased magnitude chains, good for good chains (short), good for bad chains (turn the other cheek)

How does data sampling strategy impact discovery of information diffusion in social media? (Munmun)

How can we sample social web?

Many different modes of social interaction

Scale of interest – viral marketing, ad campaigns

Is there more in social media than just scale?

Social media can have enormous power, e.g. for diffusion

However, inference of such processes is based on the quality of data

Current methods: random walk, snowball – captures structure but not content or context

RQ1: what is role of context in sampling social phenomena, RQ2: how much should we sample to capture the process

Data: Twitter. Look at diffusion via RT feature, shared URl, same hashtag

Model: Diffusion series. Has slots of individuals involved in diffusion process, links between individuals based on relationships

Sampling strategies: given N, the number of nodes to pick, topic T, social graph G

Ignoring social graph: 1. Random sampling from seeds, 2. Attribute / context sampling.

Using social graph: Forest fire (again, random or attribute)

Diffusion Saturation Metrics: user-based (volume, participation, dissemination). Topology-based: (Reach, spread, cascade instances, collection size), Time-based: rate

Our sample S distorts some metric M

Diffusion response metrics: correlate diffusion on twitter to external behavior (search and news trends)

Experimental Study

Reference Set: ~465K users, 836K edges, 30M tweets. 125 randomly chosen “trending topics” from Twitter between Oct and Nov 2009

Trending topic – theme association

Results: bias due to sampling consistent, best results come out of forest fire. For search trends, forest fire + location performs best, for news trends, forest fire + activity performs best

Larger-scale analysis: look at topic distribution by sampling strategy

Inferences about social data affected by sampling strategy. Topic + topology + seed attribute makes a difference to sampling.

Photo Tagging over Time: A Longtitudinal Study of the Role of Attention, Network Density and Motivations (Paul Russo)

Tagging over time

Overview: what factors influence users to tag?

RQs:

1. How do individual motivations affect tagging,

2. what effect does receiving attention from users affect tagging tenure

Data: Flickr – focused on established users

Individual Motivations:

-People tag for themselves (archiving, retreival) and for others (describing)

-Many ways to receive attention from others

-Huberman, Romeru, Wu demonstrated that on YuTube people whose work received more views tended to post more videos, leading to a submission cycle

-Lento et al. photo tagging behavior

New: look at network strucutre (clustering coefficient)

-Mutual Friends

H1: attention, enjoyment, commitment should increase tagging. Density should decrease tagging. ((blogger comment: what about social grooming?))

Method: 90 days of tagging on Flickr, used only “pro” users w/3 months tenure

-Combines user-reported (survey) data and system data: what people say and what they do

-Attention: comments, Density = network, motivation = survey, measured on Likert scale.

-DV = #tags/photo over 90 days

-Results: generally bear out hypotheses.

-Interesting: low density net individuals tag more than high density for same lev of attention, but higher levels of attention lead to more tagging regardless of density. On the other hand, density has positive slope w.r.t. to commitment and tagging only for low density networks, for high density networks higher levels of attention lead to less tagging for high density.

***Microblogging***

Microblogging Inside and Outside the Workplace (Ehrlich and Shami)

Method: Twitter vs. Bluetwit (internal tool)

Data: 34 users from 15 countries and 8 business units that used both Twitter and Bluetwit and 20 posts in each over 4 month period

-Twitter much more active than Bluetwit

Dataset: data collected for BlueTwit and Twitter over 4 month period. 19K posts in two tools. Extracted 4 weeks of tweets.

Manually coded 5k microblogging posts

codes: status, providing information, retweet, ask question, directed, directed q

Results:

Categories of Posts: Most frequent use – provide information, directed posts

Internal use: ask questions, directed posts, style is work-oriented

External use: provide information, style is more “social”

Public vs. Private: clear sense of what is appropriate for an internal-only audience

Reputation Management: Internal: importance of giving back

External: Publicity and Promotion

Fostering Connections: developing better awareness of professional connections in advance of a future planned or unplanned meeting

Consumption of microblogs:

-Microblogs provide early access to human selected information

Discussion:

-How is microblogging useful in a work context?

–Form of crowd-sourcing – asking questions and providing answers

–Anticipatory connections

-What differs between public and private?

–Confidentiality (very clear)

–Style of writing

–Awareness of audience knowledge and interests

–Some erosion of boundaries – despite difference lots overlap

-Why are people consuming microblogs?

–Motivation for posting – Building reputation, awareness

–Motivation for reading – early, quality news

Measuring Influence in Twitter: the Million Follower Fallacy (Cha et al.)

Goal: characterize influence in social media and study its dynamics

1. How can we measure influence of a single user?

2. Does influence of user hold across topics?

3. What behaviors make ordinary users influential?

Data: Twitter

Why Twitter? One of most popular social media, social links are primary way how information flows, traditional media soruces and word-of-mouth coexist in this environment

-54m users, 2B follow links, 1.7B links

–8.5% of profiles private

–95% users belong to the giant component

–low reciprocity (10%)

–Power law degree distribution w/extremely large hubs (500 users have more than 100K followers)

–Low tweeting activity in general (only 11% of all users posted at least 10 tweets)

Three measures of influence:

1. Indegree

2. Mentions

3. Retweets

Are these three measures related? Compared relative ranks of user across 3 measures using spearman’s rank corellation

Correlated for full population, but not for top 10% or top 1% (but retweets corr mention remains high)

top list of indegree = mix of news outlets and public figures, top retweets = celebrities, etc.

Million follower fallacy – Britney Spears has millions of followers but she doesn’t show up on top retweeted list

Measuring influence:

–Find users engage in multiple topics. Picked 3 popular topics in 2009 over 2 month period, iran election, death of  MJ, swine flu

–Focused on 13k people who talked about all 3 topics

Conclusion: just because you have a lot of followers doesn’t mean they retweet you

Along a similar idea, Tweets vs. followers in NodeXL:

Directed Closure Process in Hybrid Social-Information Networks w/Focus on Link Formation on Twitter (Romero and Kleinberg)

How do directed closed triads form?

Triadic closure vs. directed closure:

-Triadic: an edges connects 2 nodes who already have a common neighbor

-Directed: a node A links to node C to which it already has a 2-step path (through node B)

-An edge in directed graph exhibits closure if it completes a 2-step path

-Closure ratio of node C is the fraction of C’s incoming edges that exhibit closure

-could indicates how many nodes discovered C by following nodes that follow C

Data: random sample of 18 twiter micro-celebrities: Users with between 10K and 50K followers

Notation: user A is k-linked to C if A follows C and also follows k followers of C. Let s_k(C) denote set of followers k-linked to C. f(s_k(C)) = fraction of set whose edge to C exhibits closure.

Q1: Is directed closure a significant process? Randomization test

A: up to large k, f(s_k(C)) is significantly bigger than one would expect under random ordering

Observation: f(s_K(C)) increases with k, but flattens out. Why??

Properties observed: closure ratio saturates to a positive constant f, constant f is different for different micro-celebrities, constant f not closely related to total in-degree of micro-celebrity

Heuristic calculation suggests that the sum of in-degrees of incoming nodes closely predicts closure ratio

Improved Model: predict closure ratio not just by in-degree but by sum of in-degree of incoming nodes. Also sum of in-degrees of incoming nodes from same community predicts closure ratio even better!

Conclusion: definition and methodology for directed closure, evidence for directed closure on Twitter, evidence that sum of in-degrees of incoming nodes & nodes from same community predicts closure ratio on Twitter

Characterizing Microblogs with Topic Models (Ramage, Dumais and Liebling)

Do people like the posts they see?

43 users at MSR looked at 60 posts, judged as: “not really worth reading <-> maybe worth time spent reading <-> worth the time spent reading.”

the average judgment was maybe or worse, nobody on average judged posts as worth time

Fundamental problem: people followed <> Tweets worth reading

What factors go into deciding if a user is worth following?

Method: Structured interviews with heavy users, Broader survey of 56 Twitter users

Kinds of Topics for Tweets. Hobbies, professions, news, products, events = Substance. Updates about meals, travel, hygiene = status. Making plans, networking, staying in touch = Social. Humor, wit, whininess, diction, worldview = Style. Different topics liked by different users.

Shows that content is important.

Content modeling: 8.2m “Spritzer” Tweets from 2008

-Surface word features = tf idf cosine similarity, etc.

-won’t look at deeper features (e.g. parsing) – not very appropriate for Tweets

-Desparsify: Topic Models, LDA.

-Want some labels (hashtags, emoticons, questions, etc.) = Naive Bayes, SVM, etc.

-Combined surface word features, LDA, labels = labeled LDA

Content modeling with labeled LDA:

1. Discover unlabeled topics w/ k=200 latent topic dimensions (e.g. politics, sleep)

2. model common labels = 500-1000 dimensions for hashtags, emoticons, etc.

Twitter content by category: manually aggregate topics into one or more of 45 categories.

Results: 38% style, 23% social, 27% substance, 12% status

Treemap-like visualization

Filtering: Tweet stream re-ranking

Split rater’s post into train 70% and test 30%

re-rank test set by distance to positive examples

consider judgment of maybe or worth time as “positive”

Mean reciprocal Rank @ 1 Relevant: best performance = Labeled LDA + tf-idf, .75

Finding: User recommendation task

Rater’s followed user: train 6/7 followers and test 1/7. Find the test user among 8 other non-followed users. Ranking task: score by reciprocal rank of test user. Performance > .9

Next steps: better interfaces for finding and filtering and models that account for temporal dynamics

Fourth International AAAI Conference on Weblogs and Social Media (ICWSM-10)
May 23-26, 2010
George Washington University, Washington, DC

Sponsored by the Association for the Advancement of Artificial Intelligence

IMPORTANT DATES:
Tutorial Proposals: December 1, 2009
Paper Submission: January 8, 2010
Poster/Demo Submission: January 8, 2010
Paper Acceptance: March 3, 2010
Poster/Demo Acceptance: March 3, 2010
Workshop Submission: March 1, 2010
Camera Ready Copies: March 12, 2010

Featuring a keynote by:
Professor Bob Kraut, CMU,
on “Designing Online Communities from Theory”

Professor Michael Kearns, Computer and Information Science,
Univ. of Pennsylvania,
on “Behavioral Experiments in Strategic Networks”

Speakers in Special Sessions:
- Nicole Ellison, Dept. of Telecommunication,
Information Studies and Media, Michigan State Univ.
- James Pennebaker, Dept. of Psychology, Univ. of Texas, Austin
- S. Craig Watkins, Dept. of Radio, TV and Film, Univ. of Texas, Austin- Don Burke, CIA Directorate of Science and Technology, Intellipedia
- Haym Hirsh, National Science Foundation IIS Division Director
- Macon Phillips, U.S. White House, Head of New Media

Tutorial Speakers will include:
- Jake Hofman, Yahoo! Research,
“Large-scale social media analytics with Hadoop”

- Cindy Chung and James Pennebaker, Univ. Texas,
“Using LIWC to uncover social psychology in social media”

TECHNICAL AREAS
More specifically, ICWSM welcomes submissions from researchers in a number of disciplines:
- Computational Linguistics/NLP
- Text Mining/Data Mining/Machine Learning
- Psychology
- Sociology (including Social Network Analysis)
- Anthropology, Communications, Media Studies
- Visualization
- HCI
- Graph theory, concrete analysis and simulation of graphical models

Submissions are welcome that study a broad array of types social data, including:
- Weblogs, including comments
- Social Networking Sites
- Microblogs
- Wikis (wikipedia)
- Forums, usenet
- Community media sites: youtube, flickr

Technical topics of interest include:
- Psychological, personality-based and ethnographic studies of social media
- Analyzing the relationship between social media and mainstream media
- Qualitative and quantitative studies of social media
- Centrality/influence of social media publications and authors
- Ranking/relevance of blogs; web page ranking based on blogs
- Social network analysis; communities identification; expertise and authority discovery; collaborative filtering
- Trust; reputation; recommendation systems
- Human computer interaction; social media tools; navigation and visualization
- Subjectivity in textual data; sentiment analysis; polarity/opinion identification and extraction
- Text categorization; topic recognition; demographic/gender/age identification
- Trend identification and tracking; time series forecasting; measuring predictability of phenomena based on social media
- New social media applications; interfaces; interaction techniques

SUBMISSION
People interested in participating should submit through the ICWSM-10 website a technical paper (up to 8 pages, not including references),poster or demo description (up to 4 pages) by the deadlines given above (Midnight PST). Papers must be must be formatted in AAAI two-column, camera-ready style (see the AAAI author instructions page at http://www.aaai.org/Publications/Author/author.php). Details for the submission procedure will appear at the conference website: http://icwsm.org

PUBLICATION
All accepted papers and abstracts will be allocated eight (8) pages in the conference proceedings. Authors will be required to transfer copyright of their paper to AAAI.

DATA CHALLENGE
ICWSM-10 will once again hold a data challenge featuring a freely-available dataset and a half-day workshop at the conference. Details will be posted on the conference website.

CONFERENCE WEBSITE
www.icwsm.org

For general information regarding ICWSM-10, please write to icws…@aaai.org. More details about the CFP and the conference will appear on the website over time.

ORGANIZERS:
Program Chairs:
William Cohen, CMU Computer Science
Samuel Gosling, U Texas Dept of Psychology

General Chair:
Marti Hearst, UC Berkeley School of Information

Senior Program Committee Members:
Lada Adamic, Univ. of Michigan
danah boyd, Microsoft Research
Claire Cardie, Cornell Univ.
Kathleen Carley, Carnegie Mellon Univ.
Chris Diehl, Lawrence Livermore National Labs
Nicole Ellison, Dept of Telecommunication, Information Studies, Michigan State University
Lise Getoor, Univ. of Maryland
Jure Leskovec, Stanford Univ.
Winter Mason, Yahoo! Research
Kate Neiderhoffer, Dachis Corporation
Eugene Agichtein, Emory Univ.
Cindy Chung, Univ. of Texas at Austin
Scott Counts, Microsoft Research
Tim Finin, UMBC
Evgeniy Gabrilovich, Yahoo! Research
Kristina Lerman, ISI-USC
Jure Leskovec
Winter Mason
Gilad Mishne, Yahoo! Labs
Bo Pang, Yahoo! Research
Marc Smith, Connected Action Consulting Group

The Second Conference on Online Deliberation: Design, Research, and Practice (OD2005/DIAC-2005) was held at Stanford University May 20-22, 2005. From that event there is now a book,  Online Deliberation: Design, Research, and Practice, edited by Todd Davies and Seeta Peña Gangadharan (CSLI Publications, November 2009).  All content in the book is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License.

I will call out a few of the many interesting chapters, one of which I contributed to:

Chapter 5: Friends, Foes, and Fringe: Norms and Structure in Political Discussion Networks (John Kelly, Danyel Fisher, and Marc Smith, pp. 83-93)

And two from colleagues who report on tools for facilitating political debate and decision making:

Chapter 6: Searching the Net for Differences of Opinion (Warren Sack, John Kelly, and Michael Dale, pp. 95-104)

Chapter 26: Online Civic Deliberation with E-Liberate (Douglas Schuler, pp. 293-302)

The book is a great guide to the many ways computer-mediated interaction technologies are being used to build consensus or tear it apart!

—

Prof. Jen Golbeck and Justin Grimes analyzed 6,000 tweets from United States Congress members.  They found some interaction but a dominant broadcast pattern of use with a focus on self-promotion.  The Washington Post headlines the results as “Tweeting Their Own Horns Study Finds Posts By Lawmakers Boastful or Boring”

Here is the video:

From the UMD News desk:
“A new study by University of Maryland researchers finds a
growing use of Twitter among members of Congress, but that
the purpose and content of their messages fall short of
improving government transparency.

Jennifer Golbeck, assistant professor in the College of Information
Studies, Maryland’s iSchool, a doctoral student and an undergraduate
assistant analyzed more than 5,000 tweets sent by 69 members of
Congress in February. They found that House and Senate members
were using the social media platform mostly to promote themselves,
rather than engage in dialogue with constituents and the public at large.

“Members of Congress were not sharing much new information on
Twitter, and there were few posts that improve transparency,”
Golbeck says.”

Tweet the Debates: Understanding Community Annotation of Uncollected Sources

Authors: Shamma, D.A.; Kennedy, L.; Churchill, E.F.
Source: ACM Multimedia, ACM, Beijing, China (2009)
URL: http://www.acmmm09.org/
Keywords: Twitter; Debates; TV; community; multimedia; social; centrality; network

Download: wsm01a-shamma.pdf

At the core of the network are the entities that have the most “eigenvector centrality” which turn out to be the major participants in the debate.

This weekend is the Social Computing 2009 conference in Vancouver, B.C.  It is a gathering of many people doing research on social media useage.  Many papers are about tagging systems, blogs, wikis, message boards, and social networking services.

Along with several co-authors, I contributed to three papers in this year’s conference:

Bonsignore, E.M., Dunne, C., Rotman, D., Smith, M., Capone, T., Hansen, D.L. & Shneiderman, B. (2009), “First steps to NetViz Nirvana: evaluating social network analysis with NodeXL”, In SIN ’09: Proc. international symposium on Social Intelligence and Networking. IEEE Computer Society Press.

Abstract: Social Network Analysis (SNA) has evolved as a popular, standard method for modeling meaningful, often hidden structural relationships in communities. Existing SNA tools often involve extensive pre-processing or intensive programming skills that can challenge practitioners and students alike. NodeXL, an open-source template for Microsoft Excel, integrates a library of common network metrics and graph layout algorithms within the familiar spreadsheet format, offering a potentially low-barrier to-entry framework for teaching and learning SNA. We present the preliminary findings of 2 user studies of 21 graduate students who engaged in SNA using NodeXL. The majority of students, while information professionals, had little technical background or experience with SNA techniques. Six of the participants had more technical backgrounds and were chosen specifically for their experience with graph drawing and information visualization. Our primary objectives were (1) to evaluate NodeXL as an SNA tool for a broad base of users and (2) to explore methods for teaching SNA. Our complementary dual case-study format demonstrates the usability of NodeXL for a diverse set of users, and significantly, the power of a tightly integrated metrics/visualization tool to spark insight and facilitate sensemaking for students of SNA.

Abstract: Broadening adoption of social media applications within the enterprise offers a new and valuable data source for insight into the social structure of organizations. Social media applications generate networks when employees use features to create “friends” or “contact” networks, reply to messages from other users, edit the same documents as others, or mention the same or similar topics. The resulting networks can be analyzed to reveal basic insights into an organization’s structure and dynamics. The creation and analysis of sample social media network datasets is described to illustrate types of enterprise networks and considerations for their analysis.

Abstract: Community based Question and Answer systems have been promoted as web 2.0 solutions to the problem of finding expert knowledge. This promise depends on systems’ capacity to attract and sustain experts capable of offering high quality, factual answers. Content analysis of dedicated contributors’ messages in the Live QnA system found: (1) few contributors who focused on providing technical answers (2) a preponderance of attention paid to opinion and discussion, especially in non-technical threads. This paucity of experts raises an important general question: how do the social affordances of a site alter the ecology of roles found there? Using insights from recent research in online community, we generate a series of expectations about how social affordances are likely to alter the role ecology of online systems.

The 2009 American Sociological Association Annual Meeting was held in San Francisco, California, August 8-11.

The ASA attracts thousands of sociologists, a subsection of whom have  a passion for the study of the Internet and its many forms of social impacts and uses.  The Communications and Information Technology Section of the American Sociological Association (CITASA) is the group that gathers many forms of social science research on the creation and uses of information technology.  This year’s meeting included two CITASA panels, round tables, a business meeting with awards, and a (windy!) boat ride through San Francisco Bay and beneath the Golden Gate Bridge.

The CITASA sponsored papers at the conference are listed below.  The range of work illustrates the continued interest in social science studies of the impacts of information technology.

Session 1: Cultural Divergences and Convergences: Mediated Communication, Community, and Social Networks
Presider: Laura Robinson, Santa Clara University
10:30 – 12:10 PM, Location: Parc 55 Hotel

Reframing Public Space Through Digital Mobilization: The Case of Flash Mobs
Virág Molnár

The Consumption of Online News at Work
Pablo J. Boczkowski

French Software Politics: The Freedom Discourse and Globalization from Below
Sara Schoonmaker

Convergence in E-Campaigning: Comparing the Use of Attacks on German and American Political Web Sites
Eva Johanna Schweitzer

The Flow of Mediated Culture: Trends of Supply and Demand 1960-2005
W. Russell Neuman

Session 2: Sociology of Communications and IT
Presider: Keith N. Hampton, University of Pennsylvania
2:30 – 4:10 PM, Location: Parc 55 Hotel

Gauging the Impact of e-Research in the Social Sciences
Ralph Schroeder, Eric T. Meyer, University of Oxford

Does Previous Technology Use Influence Later Opinions Linking Use to Work Experiences, Relationships, and Time?
Noelle A. Chesley, University of Wisconsin-Milwaukee

Medical Conversations in Technology Enabled Communities: Perceived Benefits of Participation in Online Patient Communities
Gul Seckin

Model Failure: Assemblages, Performances, and Uneasy Collaborations in Commercial Construction
Gina Neff, University of Washington; Brittany Fiore-Silfvast, University of Washington; Carrie Sturts Dossick, University of Washington

Core Discussion Networks, Internet and Mobile Phone Use: New Media are not Increasing Privatism in America
Keith N. Hampton, University of Pennsylvania; Lauren Sessions, University of Pennsylvania; Eun Ja Her, University of Pennsylvania

————–

Round tables (1-hour): Open Submission
Organizer: Steven G. Hoffman, SUNY Buffalo
8:30 – 9:30 AM, Location: Parc 55 Hotel

Table 1: Communication and Information Technologies at the Producer-Consumer Nexus
Presider: Chris Rhomberg, Fordham University

A Tale of Two Newspaper Chains: Gannett, Knight Ridder, and the Crisis of American Newspapers
Chris Rhomberg, Fordham University

Analyzing design activity in architect and user talk-in-interaction: a preliminary analysis
Rachael Luck, University of Reading

Distributed Cognition and the Emerging Peer-to-Peer Production Model
Michael Restivo, Stony Brook University

Table 2: Communication and Information Technologies, Community, and Sexuality

Negotiating identities/remediating queering desires: Coming out and coming of age online
Mary L. Gray, Indiana University

Sociological Examination of People’s Attitude Towards Online Dating
Xue Liu, Clemson University

The Amorous Migrant: Race, Interracial Desire and Relocation in Cyberspace
Nicholas Andrew Boston, Cambridge University/Lehman College of the City University of New York

Virtually Gay: The Internet and the Construction of Hybrid Queer Identities and Transnational Gay Cultures

James Paul Thing, University of Southern California

Table 3: The Internet and Political Engagement I
Presider: Phil Howard, University of Washington

Bringing the body back in: considering the role of the body in deliberative democracy
Amy Stuart, New School for Social Research ICANN and Internet Governance
Hangwoo Lee, Chungbuk National University

Social Capital and Political Mobilization in Online Community
Jun Young Ah, Yonsei university; Jeong-han Kang, Yonsei University

Vitalizing Donation Culture in Korea: Comparison of Strategies between Traditional Off-line and New Online Donations
Sun Hyoung Lee, Yonsei University

Table 4: The Internet and Political Engagement II

A Facilitator of Civic Engagement in Online Group Contexts
Ja Hyouk Koo, University of Virginia

Muslim resistance online: a diasporic Pakistani punk music subculture on the Internet
Dhiraj Murthy, Bowdoin College

Transylvania, the Internet, and the Strategic Construction of Ethnic/National Identity
J. Patrick Williams, Nanyang Technological University

Table 5: New Technologies, Social Change and Knowledge
Presider: James A. Evans, University of Chicago

How the Internet Shrinks Knowledge By Extending It
James A. Evans, University of Chicago

Socializing the Mobile Phone: Young Urban Poor’s Fascination at Play
Lip Soon Wong, Telenor Research and Innovation Center

Social Networks inside the Enterprise: analyzing colleague link patterns in an IT workplace
Marc Smith

Table 6: Computer Mediated Communication, Social Capital, and Symbolic Interaction
Presider: Simon Gottschalk, University of Nevada- Las Vegas

Dramaturgy, Technology and Public Health: Finding Sex Partners Online among Men who have Sex with Men
Anthony P. Lombardo, University of Toronto

eIntelligence: Social Intelligence in Computer-Mediated Communication
Simon Gottschalk, University of Nevada- Las Vegas

How to Ground a Child in Cyberspace: Parents’ Exploration of Norms and Rules
Mito Akiyoshi, Senshu University

Structured Talk and Web 2.0: Blogs and Community Formation

Kenneth M. Kambara, California Lutheran University

Table 7: Communication Technologies and Network Formation
Presider: John P. Robinson, University of Maryland

Events and Attendees in Two Countries: Factors Influencing the Size and Composition of Online Social Groups
Ryan M. Acton, University of California, Irvine

From each according to media? Testing Wellman’s theory of networked individualism
Bernard Hogan, University of Toronto

The Composition of a Korean Immigrant Social Network
Sun Kyong Lee, Rutgers University

Table 8: Information Production, Coordination and Distribution

Order, Coordination and Uncertainty in Online Information Systems”
Judd Antin, University of California, Berkeley
Coye V. Cheshire, University of California, Berkeley

When High Commitment Hurts: Evidence from Wikipedia
Andreea Gorbatai, Harvard University

Wikipedia: Community or a social movement?
Piotr Konieczny, University of Pittsburgh

A new book E-Research: Transformation in Scholarly Practice edited by Nicholas W. Jankowski on the ways social science research is being changed by the rise of social media has just been released by Routledge.  My colleagues and I contributed a chapter on the ways that information visualization of social media is a useful technique to identify research questions and discover answers about the nature of human association when mediated by computation.  The volume contains work from an all-star line-up of researchers who address the opportunities and challenges of performing research with computer-mediated data about social life.

The blurb about the book describes it as:

“No less than a revolutionary transformation of the research enterprise is underway. This transformation extends beyond the natural sciences, where ‘e-research’ has become the modus operandi, and is penetrating the social sciences and humanities, sometimes with differences in accent and label. Many suggest that the very essence of scholarship in these areas is changing. The everyday procedures and practices of traditional forms of scholarship are affected by these and other features of e-research. This volume, which features renowned scholars from across the globe who are active in the social sciences and humanities, provides critical reflection on the overall emergence of e-research, particularly on its adoption and adaptation by the social sciences and humanities.”

Our chapter is “A Picture is Worth a Thousand Questions: Visualization Techniques for Social Science Discovery in Computational Spaces”, co-authored by  Howard T. Welser, Thomas Lento, Marc Smith, Eric Gleave and Itai Himelboim.  In it, we describe the ways that using information visualizations of social media data sets is a useful way of discovering insights, patterns, and clusters.  We illustrate the paper with several examples of social media information visualizations that display the range of behavior among contributors to social media spaces.

Here is the table of contents for the volume:
Introduction
1. The Context and Challenges of e-Research Nicholas W. Jankowski.

Conceptualization
2. Towards a Sociology of e-Research: Shaping Practice and Advancing Knowledge Ralph Schroeder and Jenny Fry
3. e-Research as Intervention Anne Beaulieu and Paul Wouters.

Development
4. Developing the UK-based e-Social Science Research Program Peter Halfpenny, Rob Procter, Yu-Wei Lin and Alex Voss
5. e-Research and Scholarly Community in the Humanities Paul Genoni, Helen Merrick and Michele Willson
6. The Rise of e-Science in Asia: Dreams and Realities for Social Science Research. Case Studies of Singapore and South Korea Carol Soon and Han Woo Park.

Collaboration
7. Creating Shared Understanding across Distance: Distance Collaboration across Cultures In R&D Petra Sonderegger
8. Moving from Small Science to Big Science: Social and Organizational Impediments to Large Scale Data Sharing Eric T. Meyer.

Visualization
9. Visualization in e-Social Science Mike Thelwall
10. A Picture is Worth a Thousand Questions: Visualization Techniques for Social Science Discovery in Computational Spaces Howard T. Welser, Thomas Lento, Marc Smith, Eric Gleave and Itai Himelboim.

Data Preservation and Reuse
11. Web Archiving as e-Research Steven M. Schneider, Kirsten A. Foot and Paul Wouters
12. The Promise of Data in e-Research: Many Challenges, Multiple Solutions, Diverse Outcomes Ann Zimmerman, Nathan Bos, Judy S. Olson and Gary M. Olson
13. Naming, Documenting and Contributing to e-Science Samuelle Carlson and Ben Anderson.

Access and Intellectual Property
14. Open Access to e-Research Robert Lucas and John Willinsky
15. Intellectual Property in the Context of e-Science Dan L. Burk.

Case Studies
16. Situated Innovations in e-Social Science Bridgette Wessels and Max Craglia
17. Wikipedia as Distributed Knowledge Laboratory: The Case of Neoliberalism Clifford Tatum and Michele LaFrance

JCMC Article: Discussion Catalysts

My co-authors Eric Gleave, from the University of Washington, Department of Sociology and Itai Himelboim, from the University of Georgia, Department of Communications, are pleased to note the publication of our paper “Discussion catalysts in online political discussions: Content importers and conversation starters“ in the Journal of Computer-Mediated Communication (JCMC) http://jcmc.indiana.edu/ at http://ping.fm/7NF5T

The paper describes the roles of “discussion cataylsts” who populate political web boards (newsgroups) and start the threads that get people talking!  It turns out that only a very few people in a community get to start many threads successfully.  Discussion catalysts have a knack for sparking conversations: setting the agenda for the community at large.  Discussion people have high “-in-degree”, they get replied to by lots of people, but low “out-degree”, they tend not to reply that much themselves.  The people whoreply to discussion catalysts, in contrast, do reply to one another densely.  These are the discussion people, a role  that will be the focus of a subsequent paper!

JCMC - Discussion Catalysts

Abstract:

This study addresses 3 research questions in the context of online political discussions: What is the distribution of successful topic starting practices, what characterizes the content of large thread-starting messages, and what is the source of that content? A 6-month analysis of almost 40,000 authors in 20 political Usenet newsgroups identified authors who received a disproportionate number of replies. We labeled these authors ‘‘discussion catalysts.’’ Content analysis revealed that 95 percent of discussion catalysts’ messages contained content imported from elsewhere on the web, about 2/3 from traditional news organizations. We conclude that the flow of information from the content creators to the readers and writers continues to be mediated by a few individuals who act as filters and amplifiers.

Previously, we published “Picturing Usenet” in the JCMC, a paper that features several images of information visualizations of threaded discussions and authors over time.  That paper was based on an early work with Fernanda Viegas (now at IBM Research, Cambridge, then as an MIT graduate student in the MediaLab interning with me at Microsoft Research in Redmond, Washington).
[2009 - JCMC- Discussion Catalysts - Himelboim, Gleave and Smith]

I just read a new paper from Jennifer Preece and Ben Shneiderman that provides a nice framework for the ways people contribute at different rates to collective projects in general and social media on the Internet in particular.

Preece, Jennifer and Shneiderman, Ben (2009).  The Reader-to-Leader Framework: Motivating Technology-Mediated Social Participation, AIS Transactions on Human-Computer Interaction (1) 1, pp. 13-32.

Available at: http://aisel.aisnet.org/thci/vol1/iss1/5, the paper is published in a new journal, Association for Information Systems  Transactions on Human-Computer Interaction, and is likely to be of interest to those in the social media and network analysis community. The main argument is that there are distinctive activities that people move through: initially as readers, then contributors (in small then larger ways), then collaborating with others to make larger contributions, and then to leadership (policy making, enforcement, coping with disruptions, mentoring novices, etc.).  The figure (above) from the paper is modeled on Wikipedia where these activities have been studied extensively, but they argue that these activities can be found in many technology-mediated social media.  The conversion rate from one activity to another is often as low as 1 percent (for example, there are half-a -billion readers of wikipedia, but just 1600 admins who are effectively the leaders), so the paper offers suggestions for improving the usability and sociability design to raise the conversion rate.

The Communities and Technologies conference is holding its 4th meeting in Penn State June 24-27.  This conference gathers a range of scholars interested in online community, social media, social networks, and mobile social software.  A paper “Analyzing (Social Media) Networks with NodeXL” has been accepted for publication in the conference!  Congrats to my co-authors!

Paper Title: Analyzing (Social Media) Networks with NodeXL

Authors: Marc Smith (Telligent Systems), Ben Shneiderman (University of Maryland), Natasa Milic-Frayling (Microsoft Research), Eduarda Mendes Rodrigues (Microsoft Research), Vladimir Barash (Cornell University), Cody Dunne (University of Maryland), Tony Capone (Microsoft Research), Adam Perer (University of Maryland, now at IBM Research), Eric Gleave (University of Washington)

Abstract: In this paper we present NodeXL, an extendible toolkit for network data analysis and visualization, implemented as an add-in to the Microsoft Excel 2007 spreadsheet software. We demonstrate NodeXL features through analysis of a data sample drawn from an enterprise intranet social network, discussion, and wiki. Through a sequence of steps we show how NodeXL leverages and extends the broadly used spreadsheet paradigm to support common operations in network analysis. This ranges from data import to computation of network statistics and refinement of network visualization through a selection of ready-to-use sorting, filtering, and clustering functions.

Some highlights:
An Examination of Language Use in Online Dating Profiles
Meenakshi Nagarajan, Marti Hearst

Event Detection and Tracking in Social Streams
Hassan Sayyadi, Matthew Hurst, Alexey Maykov

Gephi: An Open Source Software for Exploring and Manipulating Networks
Mathieu Bastian, Sebastian Heymann, Mathieu Jacomy

Information Diffusion in Computer Science Citation Networks
Xiaolin Shi, Belle Tseng, Lada Adamic

Considering the Sources: Comparing Linking Patterns in Usenet and Blogs
Mary McGlohon, Matthew Hurst

[Vladimir Barash is liveblogging the ICWSM conference]

10.30am A categorical model for discovering latent structure in social annotations (Said Kashoob)
Given a collection of web objects, users and tags, can we model the underlying tag generation process?

-Discover implict communities of interest?

-Categories of related tags?

-For given category, id most relevant objs for category

-compare categories

Initial thoughts: content-based topic modeling (Latent Dirichlet Allocation, LSA). Recent work applying LDA models to tags (Wu 2006, Zhou 2008)

Modeling social annotations: the process that generates content is fundamentally different from the annotation process (many authors per “document” = tag collection, not aware of each other)

Community based categorical annotation model (CCA). Communities are groups forming around interests, etc. Each community has a number of categories as its world-view. For each object, a community draws tags from the appropriate underlying categories

Object annotations are generated by communities. Each community selects tags from its category set.

-use Gibbs sampling to recover a joint distribution of tags, categories and communities

-can do inference to find most likely tags per category, per community

Content-based topics vs. tag-based categories

Exploring content vs. annotation: for pairs of objects that are similar in category space, how topically similar are they? Reslt: objects with similar content do not necessarily have similar tags and vice versa

Rubix cube example: objects similar both in category and topic are solutions to the puzzle, objects that are only similar in category are puzzles / games, objects that are only similar in topic are math pages

11am Content-based summarization and categorization in the blogosphere (Ahmed Hassan)

How can we decide which blogs are more important / influential? Given a set of blogs related to a particular topic, find a subset of blog feeds to read that have continued interest in the topic.

Can we use hyperlink popularity based algorithms for speeches and blogs? Yes, but they might not work very well

Use textual similarity to link posts instead of hyperlinks: maybe blog A affects blog B? Given a set of blogs, build a graph where nodes represent posts/feeds and edges link posts/feed with simliar text

Use a pagerank-like measure to calculate importance score of a blog in the similarity network

How can we select nodes that are important but diverse? Add discounting factor based on similarity of node to neighbors

dataset: TREC blog datase

Evaluation: use linear threshold diffusion model! How many blogs covered (activated) by first k blogs in rank. Also split data by time to see how valid is rank(t) for predicting coverage at t+1. Approach also does a little better at precision-at-k on the TREC blog dataset

11.30am Supervised ranking of syntactic configurations for finding targets of semantic expressions (Jason Kessler)

Trying to find targets of sentiment phrases (“while the dealership was friendly, the *car* was a disappointment”)

Sentiment expressions only link to physical targets (“Bill likes to drive the car”)

Two-domain corpus: cars and camera

Baselines – proximity, one-hop with dependency parser

Approach – learn to target from a corpus. Supervised ranking instead of classification. Uses linear-kernel RankSVM, off-the-shelf approach

Results – supervised ranking does better than proximity, one-hop, approaches interannotator agreement

Future work: inter-sentential target

3pm Stochastic models of user-contributory web sites (Tad Hogg)

Focus: describe aggregate group behavior

-determines structure and usefulness of user-participatory sites

Models enable:

-predicting user behaviors

-incentivizing user participation

Stochastic modeling summary:

-Start with individual user behavior, specify states and transitions between states

-Determine collective behavior (details in paper)

Illustration: Stochastic Model of Digg

-Phenomenology: users submit and vote on news stories, Digg promotes popular stories to front page, allows social networking (friends, fans)

Model of Digg voting behavior: visibility and interestingness -> votes. Extension to prior model (Lerman ’07).

- “law of surfing” for viewing web pages (Huberman et al. 98)

- incremental average growth in number of voters’ fans

- construct equation for dynamics of vote volume for a story from state diagram that formalizes visibility and interestingness. Params for vis and interest estimated from story sample. Estimate viewers watching stories from models and data.

Data: front page and upcoming stories since May 06

Modeling story visibility: story location, navigating web sites, number of fans. Each voter enables fans to see story via friends interface.

Modeling story interestingness: topic, novelty, popularity. Can estimate from web-based experiments, e.g. Salganik et al. 06, but can estimate from models and data.

Results: model captures qualitative features – slow growth initially, influence of fans on promotion, rapid growth if story promoted (much more visible to users)

Results: the number of fans have not yet seen the story drops, number of votes on story grows significantly after story gets promoted. “Promotion line” in number of fans / interestingness splits stories into will be / won’t be promoted with 95% accuracy

Predictions from early behavior: can predict #votes from first 4 votes (similar to results for YouTube), but “law of surfing” and incremental growth important parts of model

Conclusions: stochastic process approach connects user and system behaviors, applicable to social media in general when users have limited information and actions, limited use of personalized history.

3.30pm Personal information management vs resource sharing: towards a model of information behavior in social tagging systems (Markus Heckner)

Why do people tag?

Tagging: a fourth layer of indexing? (On top of author keywords, intellectual indexing by information professionals, and auto-tagging)

media type influences tagging: differences in number, language, function of tags btw Connotea, Flickr, YouTube, Delicious

Method: Scientific crowdsourcing using Mechanical Turk

Assumption: Different motivations for taggs

-Organization of one’s own digital content, i.e. personal informational management (Delicious, Connotea), vs. information sharing (Flickr, YouTube)

Questionnaire Design: Question Types

-online questionnaire posted as “human intelligence task.” asks general information, general motivation, tagging motivation and understanding, social bookmarking and search, recent usage.

Data: ~150 subjects, users of Flickr, YouTube, Delicious, Connotea

Results:

Motivation: YouTube is significantly weaker-motivated for PIM, Delicious much weaker for sharing, Flickr and Connotea about even

Perception of tagging: Connotea users perceive tagging as most easy, follow by YouTube, Delicious, Flickr (not significant). Connotea user agree very strongly that tagging is a useful feature

Towards a model of tagging behavior: Shneiderman’s approach towards social software (social spheres), etc.

4pm Motivational, structural and tenure factors that impact online community photo sharing (Oded Nov)

Why do people in online communities share? Can we quantify the drivers for sharing (or not sharing) and their effect on actual behavior?

Three types of questions as framework:

Why – drivers of sharing (Motivation, structural properties, personality, privacy concerns)

What – type of information shared (code, content/facts, meta-info (tags), photos)

Where – context of sharing (OSS, Wikipedia, Flickr)

Creation vs. sharing: the act of sharing is separate from the act of creation

People take photos regardless of the sharing act (really?), the “second act” of sharing photos is optional, separate from the “first act” of photo sharing

Identifying the factors in sharing: motivational (extrinsic vs. intrinsic), structural factors (position of user in community network), tenure in community

Motivations: enjoyment (self/intrinsic), commitment to the community (others/intrinsic), self-development (self/extrinsic), reputation (others/extrinsic)

Response variable: artifact sharing per tenure year, IV: motivational vars + structural (number of contacts) + tenure (years since started sharing)

Method: combine user-reported (survey) data and system data: what people say + what people do. N=278, used only “pro” users (>200 photos) with at least 3 months’ tenure on Flickr.

Results: significant positive effect of commitment, negative of self-development, positive of number of contacts, negative of tenure, rest not significant.

Why is enjoyment not correlated with sharing? Users may be motivated more by “fun” of creation rather than content sharing.

Why is correlation between self-development and photo sharing nefative? A tradeoff between contribution quality and quantity? Greater self-development motivation -> focus on the quality of artifacts shared, at the expense of quantity

Summary:

quality / quantity tradeoff, fun is not an issue?, diminishing sharing

4.30pm Modeling Blog Dynamics (Michaela Goetz)

Blogosphere is a system of interactions: Entities: Bloggers, Posts, Topics

Model: simple set of rules (followed by blogger) that creates these interactions

Evaluation: creating a synthetic blogosphere, comparing it to real blogosphere

Motivation: forecasting, advertising

How is this different from modeling social network? 2 networks combined: Blog vs. Post network, complex temporal dynamics

goal: model micro-level interactions to observe macro-level interactions in blogosphere

Properties of the blogosphere:

-Topological – blog, post = follow power law distribution

-Temporal – user posting activity, popularity over time (link creation)

Burstiness (Slope = 1 of aggregation level vs. entropy) & Self-similarity (Linearity of aggregation level vs. entropy)

Inter-posting time follows a power law

Time t vs. number of in-links t days after publishing follows a power law

Desired model: simple (no parameters), intuitive (local rules), creates realistic topology and dynamics

First-try solution:

-inter posting times sampled from exponential distribution, links created using pref attachment – leads to exponential inter-posting distribution, poisson degree distribution (really?), etc.

Second-try solution (Zero-Cost):

In every round, for every blog, user u takes a random walk step, if he reaches 0, he decides to post P

when he posts, he can make a link or not, if he makes a link, he can choose a neighbor based on frequency of links or non-neighbor, then chooses some post of neighbor and links to random posts upward in the cascade

This model accurately reproduces both the topological and temporal patterns (at a qualitative level – same distributions, different though relatively close exponents. Biggest difference: 1.5(sim) vs. 0.7(real) in inter-posting time exponent)

[Vladimir Barash is liveblogging the ICWSM conference]
9-10AM: A Tempest: Or, on the Flood of Interest in Sentiment Analysis, Opinion Mining, and the Computational Treatment of Subjective Language (Lillian Lee)

-Sentiment analysis using discussion structure: clasify speeches in US congressional floor debates as supporting or opposing proposed legislation -Individual doc classifier -agreement (degree) classifier for pairs of speeches

-Agreement info allows COLLECTIVE CLASSIFICATION – “agreeing speeches should get the same label”

-ECON: debate about effect of sentiment on sales
-comScore (users willing to pay 20-99% more for 5 star item vs. 4 star item)
-Jury is still out

-SOC: What opinions are influential? (Niculescu-Danescu Muzyl et al.)
-Prior work has focused on features of text and has not been in context of sociological aspects of reviews
-look at helpfulness scores

-What about review’s star rating in relationship to others?

-theories from soc / social psych:
-conformity
-brilliant but cruel

-Are the social effects just textual correlates?

-would like to control for actual quality of review text. Manual annotation? Tedious, subjective. Automatic clasification? Need extremely high accuracy guarantees.

-use plagiarism (1% of all reviews) to control for text quality! findings hold for plagiarized pairs

Summarizing:

-Sentiment analysis has many important applications – to researchers, to citizens, to governments

-encompasses many interesting research questions

-extends to many areas

Stand-out question: matt hurst and the user as generative model for opinions

10.30 AM: Gesundheit! Modeling Contagion through Facebook News Feed
(Eric Sun, Itamar Rosenn, Cameron Marlow, Thomas Lento)

Motivation: how do ideas diffuse through a large social network?

-Theory of the Influentials (Gladwell)

-Accidental Influencers(Watts): anyone can be an “influencer.” Ideas don’t spread via influentials, ideas spread like viruses (susceptible or not), goal to find a large number of susceptible people

Q: are contagions triggered by small # of sources? What are some characteristics of diffusion chains on Facebook? Can we use demographic or behavioral characteristics to predict size of diffusion chains a particular user will create?

Spreading ideas on Facebook – through News Feed

-Page Fanning = becoming fan of people, orgs, etc.

-Mechanics: Alice fans a page, Bob sees Alice’s action on his News feed, Bob fans page as well (link: Alice -> Bob)

-Strong ties: links depend both on friendship and on actions (following)

-Median page has most of its fans in one (weakly) connected cluster

-Large clusters Not Started by “one guy” – roughly 15% of fans in the biggest cluster of each Page are start points

-Variability in this percentage becomes very small as #fans increases

-Clusters are formed when many short diffusion chains merge

-Data: actor to follower connections for ~300,000 FB paes

-Main dataset: page-level data

-Second dataset: select 10 random, representative pages (at least 40 days old had at least 5k fans) and analyze users that start chains

-Prediction Model: Response = max_chain_length, Predictors: gender, log age, log FB age, etc. Method: 0-inflated neg binomial regression

-results: Demographic characteristics not important, number of Facebook friends not important, feed exposure is the strongest predictor with coefficient ~ 1 (so a 1% increase in the number of people who see ego’s fanning ~ 1% increase in chain length)

-Comment: this is global focus, not local focus. What about the interpersonal dimension, i.e. the likelihood that Alice infects Bob?

-Comment: support for Duncan Watts’ idea

————————————————————————————————————————————————-

11am Seeking and Offering Expertise across Categories: A Sustainable Mechanism Works for Baidu Knows
(Jiang Yang, Xiao Wei)

Baidu Knows: Chinese QnA site

-Growing extremely fast: more than 80 million questions asked in 4 years.

-Huge user population (2.6 mln users). Knowledge repository as online source

-Points! Points! Points! (flexible amount of extra points set for best answerer, more points buy more answers, etc.)

-Building sense of community: honor title system (including traditional Chinese titles!), online chats, etc.

-Data: Full history of QnA 12/07-05/08. 9.3 mln questions asked, 5.2 mln (56%) resolve, 2.6 mln users participated

-3.3 answers for each question (vs. 7.3 for Yahoo! Answers, note that Yahoo! Answers encourages answering more than asking)

-Significant categorical difference in awarded points: low(brand, science, food) vs. high (medicine, computer, music)

-Price of answering positively correlated to popularity of category

-Order difference: according to human rating of sample questions, order of answers doesn’t matter, but first answer has highest chance to be best answer, more points awarded for later best answers

-Reinforcement cycle: encourage continuing

-Answerer performance positively correlated with activity level. More active answerers choose less expensive questions, questions with fewer answers. More active answerers working harder (longer answers), and more focused (on particular category)

Reinforcement cycle: choose less competitive q’s -> better performance -> more efforts -> more focused -> choose less competitive q’s

-Askers: learn how to better ask: more active askers, ask cheaper questions, experienced askers get more answers with per point they ask (slight trend).

-Asker/Answerer hybrids (22% of pop): core of contribution! Much more active (almost 1/2 total questions), more generous (offer higher award: 12.3 per question versus 11.6 on average in general, share same pattern as normal asker but paying higher each time), not necessarily experts, incentivized

Seeking and offering across categories: some categories are pretty self-contained, others are more porous. Lots of cross-category contribution

-A sustainable mechanism is working on Baidu Knows (that’s a good discussion question!)

11.30 AM: Community Structure and Information Flow in Usenet: Improving Analysis with a Thread Ownership Model (Mary McGlohon, Matthew Hurst)

-Compare communities of online social nets may lend insight into how groups form and thrive

-How does info diffuse between communities?

Data: Usenet, 200 politically-oriented newsgroups (bulletin boards) – polit in name, Jan 04 – june 08. several countries, 19.6 mln unique articles, 6.2 mln cross-posted

Cross-posting: large % of articles are cross-posted to multiple groups. Somebody reading one group may “reply-to-all” such that all groups see it.

Structural analysis: how do edges btw authors form? How does the reciprocity of groups compare? How can we measure similarity btw groups?

-Make network of authors for each group, if a_1 has replied to a_2 at any point, there is an edge from a_1 to a_2. Find power law relationship btw #of nodes and #edges over time (similar to Leskovec et al. densification). Exception: tw.bbs

-Reciprocity: which groups have highest reciprocity? Top 10 were European newsgroups, e.g. hun.politika (up to .58). Lowest reciprocity: tw.bbs

-Similarity: use Jaccard coefficient for cross-posts = #shared articles btw 2 groups / Total # articles in groups, can do same with shared authors

Highest similarity ~.54 (bc.politics and on.politics).

Draw thresholded similarity network, find clusters: parties, US regional, countries, alt.politics subgroups

-Image: english-speaking countries cluster. Can.politics (Canada) highly central!

Ownership Model: we would like to find out in which group the activity is truly occurring. How can we trace this? ANswer: assign “ownership” based on authors of posts. First, assign authors to groups based on devotion, where devotion(a,g): what % of an author a’s posts are exclusively posted to a given group g

-For all groups that author posts particular post p to, the post belongs to the group with the highest (normalized) ownership between it and the author

-Example: “Kiss the National Parks Good-Bye” initially corss-posted to several groups, 38 groups in total, ownership concentrated in seattle.politics and or.politics

Information flow between groups: How often does an author in group 1 respond to a post in group 2? Define influence g_a, g_b as the product of the groups’ devotion scores for a particular author

Ownership-based similarity. Q: How can ownership help us more precisely state when 2 groups are similar? Use devotion instead of Jaccard to calc similarity between groups

-Potential applications: link prediction, IR and relevance, ownership for email lists. Future work: use ownership to predict whether group will continue or die off

1.30pm Does Showing off Help to Make Friends? (Christophe Aguiton)

Self exposition and social capital:

-What do we let others see about ourselves on social networking sites?

-How do we relate to others depending on what they show?

Game sociological survey: link

part 1: if you were on these pics, which would you publish on a website?

part 2: questionnaire

part 3: down-to-one-friend (start with x friends, see pics only, eliminate one; add favorites info, elminate one; etc. down to one friend)

first launch: FB, diffuses to blogs, Flickr, news, 15,000 respondents by end of experiment.

-Sample is not representative of French SNS users. Lots of heavy internet users. 71.1% male, average age 28 years old, 47% high school diploma, 33% students.

First Results:

-moderate / controlled level of exposure (exposure score: ~2.4 on 1-4 scale)

-extraversion index, socializing index

Method: PCA to cluster photos in the experimental dataset. Four components: traditional self-exhibition (ordinary life situation), bodily immodesty (nudity / sexual situations), showing off (protests, etc.), provocative (negative activity).

-Cluster analysis with scores of PCA, five clusters: Modest (people don’t like to show themselves, 19%, more women, older, high level of ed, high status position, few friends) + four from above.

-Main question: find no correlation between sns use and level of self-exposition

-2nd question: how do people make friends?

- popular friendship targets (from 3rd part of game) are young, cool, active, unpopular are  older, more reserved

- subject choice largely guided by homophily, a tendency to bond with similar others. Results: people preferentially choose as friends of same age and diploma level. Heterophily by gender: both men and women choose women over men.

- What aspects of persona do different kinds of people look at? Modest people most closely look at “about me,” traditional exhibitionists most closely look at “wall,” provocative most closely look at “sexual preferences”

Main results of survey:

-Self-exposition on web is a social construction, requires reflexive and strategic control of one’s image, self-exhibition strategies differ according to sociological factors, social networks encourage homophily but also allow users to have more heterogeneous social capital

2pm. What are they blogging about? Personality, topic and motivation in blogs (Alastair J. Gill et al.)

How does personality influence blogger motivation?

Personality – describes fundamental core of individuals

-Behavior and preferences

-Useful for categorising users and consumers

-How does this influence bloggers? Blogs – unique freedom of expression for authors

-Already shown to influence langauge in CMC (Gill 2004, Nowson 2006).

-Analysis of Polish blogs w/ suggested psychological profiles)

Motivations: Internal – Documenting life, catharsis (therapy); External using own perspective – Interests, Opinions

Personality: Big Five model of personality (Goldberg ’92, Costa and McCrae ’92).

Data and Method: Internet meme personality test: 5 Y/N questions each for the Big Five personality types -> high-mid-low scores; 3 months of blogs extracted from Nielsen BuzzMetrics data. Basic statistics, text analysis.

results:

Neuroticism: use of blogs for self-therapy/catharsis – focusing on self and venting purely negative feelings

Extraversion: life narrative (documentation) in conversation with reader; expressing highs and lows, but not mundane. Use of 2nd person pronouns

Openness: review or evaluation of leisure (music, TV) from personal perspective, but no increase in thinking or senses

Conscientiousness: faithfully document life going on; references to others; positive emotion. Job focus, little temporal narrative.

Agreeableness: positive self-talk focus

Discussion: Blogs unsurprising mainly focus on self. Face apparently genuine in blogs. Agreeable bloggers provide a barometer of what is / isn’t acceptable in blogs

2.30pm A social identity approach to identify familiar strangers in a social network (Nitin Agrawal)

Who are familiar strangers?

Observe repeatedly, but do not know each other: Real world – people you see daily on a train (going to same workplace); Blogosphere – people who have similar blogging behavior / interests but not in each other’s social networks

Together, familiar strangers form a critical mass: understanding of one blogger gives a sensible and representative glimpse to others -> better customization, personalization and recommendation.

Familiar strangers in social media: an example, u is a blogger with interests A_u, friends v_1… v_k with interests A_v_1… A_v_k. Find non-adjacent u’ with similar interests (intersection of A_u, A_u’ is non-empty).

-Egocentric network view (exposure to network limited to neighbors).

-Social identity approach: cluster contacts into groups, propagate search through relevant clusters of contacts (prunes search space). For this to work, network needs to be a small world (WS 98)

-Method: represent contact by tag vector, content vector, use cosine similarity, then k-means clustering

-Ground truth: Global network view. Data: Blogcatalog (~24k nodes), DBLP (~35k nodes). Also compare to exhaustive and random search strategies.

Results: 79.3%+-3 for BlogCatalog, 91.3%+-2.1 for DBLP, greatly reduced search space.

3pm You are where you edit: Locating Wikipedia Contributors through Edit histories (Michael Lieberman, Jimmy Lin)

Minig Wikipedia: id Wikipedia contributors who edit geopages in a constrained space, have specific “pet” geopages (pages for geographical locations identified with geotags)

Features with extent: all geopages tagged with single lat/lon, even though they can be countries, cities, rivers, etc.

Wikiepdia edit histories: ignored anon edits, minor edits, focused on edits to geopages

Edit area = convex hull of geotags smaller than 1 degree sq. Account for outliers with simple approximator that cuts off at F closest-together geotags

Results: Pet Geopages. Over 50% of contributors with 5-20 edits, and 25% of contributors with over 20 edits, have 80% of edits to 1 or 2 geopages

Reasons for Tight Edit areas: randomly selected 100 contributors with at least 10 edits to geopages and small edit areas. Concurrently examined contributors’ user pages and the set of edited geopages to determine an interest. Contributors with small edit areas tend to be born in or are living in close-to-edit areas.

Future work: using alterante measures to determine geopage edit significance

4pm CourseRank: a closed-community social system through the magnifying glass(Georgia Koutrika)

CourseRank: community for Stanford students to evaluate courses, browse courses, plan academic program, interact with each other, ask / answer questions. 1.5 years, 11k students, 19k courses, 3k reviews

Special features: well-defined closed community, multiple constituencies (staff, students), special-purpose tools, hybrid data

A new class of social sites defined by these characteristics. E.g. university social site, scientific social site, A-space (intelligence)

Popularity: >85% of Stanford students are CourseRank users

Usage: follows academic cycle

Participation inequality: 20% created by intermittent, 80% by active; 31% of lurkers, 38% intermittent, 30% (!) active

Smaller communities (departments) breed more active students

Truths and Lies: grade distribution follows official. Good incentives make better users (is this really evidence?). But there is bias: correlation between grade given to student and rating given by student

Lessons Learned:

-added-value services a big thing

-high-quality data

-community feeling is strong = students coming together with common need

-meaningful incentives

4.30pm Using transactional information to predict link strength in online social networks (Indika Kahanda)

OSNs (Online Social Networks) are larger and more heterogeneous than manually-collected social networks

High median degree implies presence of many weak links

Conjecture: Link strength can be predicted from transactional information

Data: Purdue FB. Transactional info: Wall comm, photo postings, group memberships. Networks over Wall, Pictures look more like offline-collected networks (e.g. AdHealth data)

Automatically identifying top friends: link strength prediction task (binary)

Related to, but different from, link prediction (which focuses on predicting future links between u,v in a unimodal network). Previous approaches use attribute similarity features or topological features of network. Adamic and Adar (’03) used ancillary networks but focused on similarity vs. transaction

Feature types: Attribute-based (attribute similarity btw two nodes), Topological features (assess connectivity of users in friendship network), transactional features (number of bi-directional wall/photo/group posts), network-transactional features (assess connectivity of users in transaction networks)

Experiment 1: Feature rankings. Compare relative importance of each of 50 features, using info gain and chi-square statistic. 12 of top 15 are network-transactional features, 3 are transactional, 12 use wall info, 3 use picture info.

Experiment 2: Feature type comparison. Ablation study. Network-transactional features achieve best performance

Experiment 3: Link type comparison. Ablation study using data from each link type separately (all features). Wall information results in best performance. Picture info does not improve performance because of sparsity

Experiment 4: overall classification results. Bagged decision trees perform best.

Results indicate that transactional events useful for presenting link strength, but should be used in context of larger network for best performance

5pm RevRank: a fully unsupervised algorithm for selecting the most helpful book reviews (Oren Tsur)

Most reviews are: repetitive, limited contribution, poorly written, unnoticed

User voting bias: Liu et al. – imbalance vote bias, early bird bias, winner circle bias. Many very helpful reviews go unnoticed.

Interesting features of reviews:

-there are a lot of them

-contributors put big cognitive effort to generate them

-Good faith. Reviewers expect no direct reward.

Main idea: automatic detection of dominant concepts. Dominant concepts are either really frequent or infrequent but very informative. Term dominance defined as ratio of term frequency in review set to term frequency in balanced review set (British National Corpus)

RevRank algorithm: find most dominant concept, vectorize, rank reviews according from centroid identified by the core vector

Experimental setup: 12k reviews for Da Vinci Doe, World is Flat, Harry Potter, Ender’s Game. Compared to random, user votes. Gold standard – human labels.

Results: in 85% of test batches, RevRank pick was ranked “the most helpful.” In some cases, random algorithm outperformed user votes!

Summary: RevRank is fully unsupervised, better than user votes, finds “hidden” reviews and interesting insights

End of Day 1

On Tuesday night Marc Smith and I will be presenting the poster for our paper, “Distinguishing Knowledge vs Social Capital in Social Media with Roles and Context” at the International Conference for Weblogs and Social Media. You can find the poster, co-authored with Marc Smith (Telligent Systems), Lise Getoor (University of Maryland) and Howard T. Welser (Ohio University), here. The full text of the paper has more information, but the poster is a good summary of the key concepts in the paper:

• What roles do people play in social media?
• What contexts shape user behavior in social media?
• How can we leverage roles and context together to predict future user behavior (in terms of contribution type) from past user behavior?

The specific research question being addressed is: can we predict whether a particular contribution to Live Q&A (a Microsoft-sponsored community question answering site) will be contain factual information, or discussion / chat. It is possible to do the prediction based on the text of the contribution, but such an approach focuses entirely on the content, and not on the actor – the user who is making the contribution. If we leverage actor-centric information (what role does he/she play in the community: an “answer person” or a “discussion person”? is he making the contribution in a discussion-oriented context, such as implied by tagging the contribution as “fun,” or a fact-oriented context, such as implied by tagging the contribution as “math”?), we find we can build a decent predictor at very low cost with very few variables. If we use just role information or just context information, we do reasonably well… but if we use both, we do *much* better. While the question we’re answering here is quite specific, the advantage of our approach is that it can be applied to almost any social media context – any place online where users can both contribute content and interact with others. We could just as easily create a predictor for contributions to Yahoo! Answers, or even to Wikipedia (if we had the relevant data). This is definitely food for thought / opportunity for future work

Our (Vladimir D. Barash, Marc Smith, Lise Getoor, Howard T. Welser ) poster paper, Distinguishing Knowledge vs Social Capital in Social Media with Roles and Context  has been accepted and published at the 2009 ICWSM (International Conference on Weblogs and Social Media) conference which will be held in San Jose, California this May 17, 2009 – May 20, 2009.

Abstract
Social media communities (e.g. Wikipedia, Flickr, Live Q&A) give rise to distinct types of content, foremost among which are relational content (discussion, chat) and factual content (answering questions, problem-solving). Both users and researchers are increasingly interested in developing strategies that can rapidly distinguish these types of content. While many text-based and structural strategies are possible, we extend two bodies of research that show how social context, and the social roles of answerers can predict content type.  We test our framework on a dataset of manually labeled contributions to Microsoft’s Live Q&A and find that it reliably extracts factual and relational messages from the data.

Full Text: PDF: 2009 ICWSM Distingusihing knowledge versus social capital

Princeton Center for Information Technology Policy

I attended the “Studying Society In A Digital World” conference at the Center for Information Technology Policy at Princeton University.  They just posted most of the conference slides.  I took some pictures and have inserted them next to the link for slides where I had a picture (or a good one!).  The conference was very useful and informative: there is a great trend towards sensor driven data sets that, in aggregate, illuminate large complex systems in detailed and surprising ways.

Talks from SenseNetworks and MIT made the vision of a continuous “trail” document assembled by location and biological sensors from every human on earth seem not so outlandish. Samuel Madden from MIT spoke about opportunistic mobile wifi connectivity in moving vehicles.  MIT rebuilt the WiFi stack to enable 13ms associations instead of 13 second associations with an access point.  The result is that a car with such a WiFi card can drive along Boston city streets and exchange about 200KB a minute with open unsecured access points along the way.  Free bandwidth in the city.  What do they do with it?  They stream live telemetry of a fleet of cabs.  The cabs have accelerometers on them and GPS which is reported in almost real time back to a server.  Along with the engine computer’s data, they collect a ton of data about traffic and road surface quality.  They can see changing patterns in the activity levels of the cabs and infer changing activity at businesses.

A major theme of several presentations was crowdsourcing for science, with talks about ebird.org and galaxyzoo highlighting a distinction between sites that enable a group to collect data (ebird) – with the associated issues of data validity — and those sites that enable a group to annotate data (galaxyzoo) that has already been expertly collected.

Lada Adamic: Social Influence and the Diffusion of User-Generated Content

Chris Barrett: Co-evolution of Sociotechnical Networks and Individual Behavior

Kathleen M. Carley: Information & Belief Diffusion Through Social Networks: Empirically Grounded Simulation

Damon Centola: New Theory and Experiments on Diffusion in Social Networks

Pablo Chavez: The Current Policy Debates Over Online Information Practices: Implications for Research in the Digital Age

Nosh Contractor: Digital Traces: An Exploratorium for Understanding & Enabling Social Networks

Nathan Eagle, Michael Macy: Scaling of Sociodynamics

Scott Golder: Temporal Rhythms in Electronic Society: Examples from Facebook and Elsewhere

Eric Horvitz: Through the Lens of a Large Instant-Messaging Network: Planetary-Scale Views on Behavior

Tony Jebara: Learning Networks of Places and People from Location Data

Steve Kelling: eBird: The Long Tail of Community Engagement in the Scientific Process

Jon Kleinberg: Spatial Signatures of On-line Behavior

Robert Kraut: Theory-Based Design of On-Line Worlds

W. Russell Neuman: Social Science and Policy Praxis

Jukka-Pekka Onnela: Using Cell Phones to Study the Large-Scale Structure of Social Networks

Paul Resnick: Understanding Opinion Diversity Preferences Through Field Experiments

Matthew Salganik: Community-Generated and Community-Sorted Information In his presentation Matt made the remarkable connection between deliberative democracy and the cat comparrison site: Kitten Wars.  His talk introduced a model for a kind of Am I Hot or Not for political discussions.  His group built a web site that helped the student community at Princeton set its priorities for student government.  The work has significant implications fo deliberation tools for organizations and enterprises.  Unlike systems that simply encourage users to contribute ideas to a potentially long and never acted upon list, this system forces a comparison task that can be performed in one click but demands implicit contrasts and estimation of value.  The use of the almost adictive “hot or not” style interface (or more accurately, kittenwars)  allows users to decide between, for example, longer hours for the student cafeteria or expanded video rental services, and get presented with their estimate in the context of other’s choices and the opportunity to choose between two things again.  After a population has run through a set of pair-wise contrasts a broader sense of the priorities of the community can be calculated.

In my talk, I focused on the idea that information want not to be free or expensive, rather, information wants to be copied.  Like DNA, the goal of any string of bits is to make a duplicate copy of themselves.  Several technical realities mean that while information may exist on a spectrum from private to public, it only moves in one direction (public) and almost never back.  Once made public on the Internet, even if only for a moment, a photo, document, or other digital object is almost certainly to have been copied, indexed, backed up, or replicated.  All efforts to delete a digital object once widely distributed is like trying to take wine out of water.  This is because all cryptography become brittle over time, most bits end up exposed after they get distributed, and more events trigger widespread distribution of bits than expected (for example, linking a photo, and a location, to a tweet that gets copied to LinkedIn and Facebook, that then appears in an RSS feed and is copied from there to Friend Feed.  As it travels, information looses more of the access controls that initially made it relatively private until it is effectively public.

Marc Smith talks about information at Princeton CITP "Studying Society in a Digital World" Conference - Photo Credit: Scott Golder

Marc Smith: Autobiography, Mobile Social Life-Lagging and the Transition from Ephemeral to Archival Society

Joshua Tauberer: Watching the Watchers: Government Oversight with Civic Hacking

Marshall Van Alstyne: Information, Social Networks and Productivity

Sadly, no picture for Luis Van Ahn’s talk: however, this presentation was a fascinating review of the capcha and re-capcha services and the new direction of providing translation services as language learning games.  Luis Van Ahn invented capcha, felt bad about the cumulative human time wasted by filling out those squiggle word puzzels to get on a web site, and decided to harness capchas to a useful task: text recognition for books.  To translate words from bad scans of books that the OCR software fails to recognize correctly, the garbled data is presented to humans, who, collectively, have translated millions of previously unintelligible words.  Now, his new project is to expand the small user population of bi or multilingual speakers who can translate between languages.  The approach applies the “Mechanical Turk” “human intelligence task” concept to language translation.  His language translation service presents foreign language sentences to users with all dictionary words from a simple translation listed below.  Users click on best word selection beneath each foreign word.  The surprising results: pretty good translations AND users start learning a foreign language!

Luis von Ahn: Human Computation

http://www.ncess.ac.uk/conference-09/

5th International Conference on e-Social Science

24th – 26th June 2009

Here is a conference for those in Europe: a meeting at the Maternushaus in Cologne, Germany is the 5th International Conference on e-Social Science.  The conference goals:

The aim of the annual international conference on e-Social Science is to bring together leading representatives of the social science, e-Infrastructure, cyberinfrastructure and e-Research communities in order to improve mutual awareness and promote coordinated activities to accelerate research, development and deployment of powerful, new methods and tools for the social sciences and beyond.

We invite contributions from members of the social science, e-Infrastructure, cyberinfrastructure and e-Research communities with experience of, or interests in:

* exploring, developing, and applying new methods, practices, and tools afforded by new infrastructure technologies – such as the Grid and Web 2.0 – in order to further social science research; and

* studying issues impacting on the wider take-up of e-Research.

Another conference focused on research on blogs and other forms of social media is “ICWSM” – the International Conference on Weblogs and Social Media.  I was able to attend the previous meeting of this conference last March in Seattle and give a talk about different classifications of social media and I am looking forward to attending this year’s meeting in San Jose.  Last year we had a poster paper in the conference about the ways some users in a blog system called Wallop were able to hold other users in the system.

Some Users Pack a Wallop: Measuring the Impact of Core Users on the Participation of Others in Online Social Systems
Thomas M. Lento, Eric Gleave, Marc A. Smith, Howard T. Welser

There was also a paper about the lessons learned from managing large corporate online community efforts.

Space Planning for Online Community
Danyel Fisher, Tammara Combs Turner, Marc A. Smith

This year, we have a poster in the conference that is focused on the ways network structures created when people reply to one another can be used to predict whether a message or thread is a question and answer exchange or a long discussion or debate.

Distinguishing Knowledge vs. Social Capital in Social Media with Roles and Context
Vladimir Barash, Marc Smith, Lise Getoor, Howard Welser

The conference attracts some great people and features the state of the art in research at the intersections of computer science, natural language processing, social network analysis, search engine/information retrieval design, information visualization, knowledge management and the social sciences.  That can be eclectic but this is the place for hearing about new work on Wikis, Blogs, Message Boards, and other social media systems like social networking services, micro-blogging systems, and mobile software.

The conference is held this year in May, from the 17th-20th, in San Jose, California.

Here are my pictures from last year’s ICWSM in 2008, held in Seattle, Washington.

There is also a nice picture from Joe McCarthy of Tom Lento and me in front of our poster at ICWSM 2008.

by gumption

2009 Communities and Technologies Conference

C&T 2009
June 25-27, 2009
The Pennsylvania State University
Information Sciences and Technology Building

In an increasingly networked world, the concept of community has taken on new meanings and inspired the development of a wide range of technologies aimed at forging connections, improving communication, and enabling coordination among groups of people. Today, such terms as virtual community, blogging, podcasting, and smart mobs have become commonplace, yet each represents a complex system of hardware, software, and people, shaped by perceptions, norms, rules, and habits, and occurring within varied social and cultural settings.

The Communities and Technologies biennial international conference serves as a forum for stimulating and disseminating research on the complex connections between communities – both physical and virtual – and information and communication technologies. Researchers studying aspects of this interaction between communities and technologies, regardless of disciplinary background, are providing original contributions to the Fourth International Conference on Communities and Technologies.

http://cct2009.ist.psu.edu/workshops.cfm

http://cct2009.ist.psu.edu/program.cfm

I am please to note that a paper about NodeXL will appear at the Communities and Technologies conference, there will also be a workshop on the use of the tool presented prior to the official start of the conference.

Penn State hosts C&T 2009

Hope to see you there!

Here are some pictures from the prior Communities and Technologies conference in 2007 on the campus of Michigan State University in East Lansing, Michigan.

2009 Social Computing Conference

A great conference on social media and related issues will be held this August in Vancouver: 2009 IEEE Social Computing Conference.  I will be reviewing for the conference and will be working on a couple of papers that will be submitted for review.

http://cse.stfx.ca/~socialcom09/

Topics of particular interest for the conference include, but are not limited to:

· Social Computing Theories
· Social System Design and Architectures
· Social Network Analysis and Mining, Semantic web
· Social Behavior Modeling
· Social Intelligence, Social Cognition
· Social Computing Applications such as collaborative filtering, bookmaking, tagging, and multi-agent systems
· Human Machine Interactions
· Emotional Intelligence, Cultural Dynamics, Opinion Representation, Influence Process
· Data Mining, Machine Learning, Information Retrieval, Artificial Intelligence in Social Contexts
· Trust, Privacy, Risk and Security in Social Contexts
· Social Engineering, Tools, and Case Studies
· Services Science, Quality, Architecture, Management, and Tools
IMPORTANT DATES
Workshop Proposal: March 01, 2009
Paper Submission Deadline: April 15, 2009
Authors Notification: June 01, 2009
Final Manuscript Due: June 15, 2009

Professor Ben Shneiderman, from the University of Maryland, Department of Computer Science, will deliver one of the keynote addresses on: “Network Analysis and Visualization for Understanding Social Computing”.

Hyperlinked Society

In my paper From Hyperlinks to Hyperties in  The Hyperlinked Society: Questioning Connections in the Digital Age (Joseph Turow and Lokman Tsui, Editors; which came from the symposium of the same name at Annenberg School of Communication in Philadelphia).  I propose one of the ways mobile devices and social software will change the nature of the “interaction order” is to create “hyperties” – here is the abstract of the paper:

With a  very nice letter that puts the award in some context:

Ten papers out of 515 at the conference were selected for Best Paper Awards.  Many thanks to track organizers Karrie Karahalios and Fernanda Viegas.

A previous paper in 2006 also got best paper: You Are Who You Talk To: Detecting Roles in Usenet Newsgroups, by Danyel Fisher, Marc Smith, and Howard T. Welser

Two years before that in 2004 Fernanda Viegas and I also published a paper at HICSS that got best paper: Newsgroup Crowds and AuthorLines: Visualizing the Activity of Individuals in Conversational Cyberspaces,
by Ferndanda B. Viégas and Marc Smith.

Tom Erickson maintains a great listing of many years of HICSS papers.

Here is Tom Lento receiving the award at the conference in Hawaii earlier this month:

More Best Papers from this year…

	Collaboration Systems and Technologies Track Minitrack: Cross-Organizational and Cross-Border IS/IT Collaboration Paper Title: Breaching the Knowledge Transfer Blockade in IT Offshore Outsourcing Projects – A Case from the Financial Services Industry Authors: Robert Gregory, Roman Beck, and Michael Prifling
	Decision Technologies and Service Sciences Minitrack: Information Systems for Sustainable Development Paper Title: Sustainability of Grasshopper Management and Support through CARMA Authors: John Hastings, Alexandre Latchininsky, and Scott Schell
	Digital Media: Content and Communication Minitrack: Social Spaces: Production and Consumption of Goods in Digital Collectives Paper Title: A Conceptual and Operational Definition of ‘Social Role’ in Online Community Authors: Eric Gleave, Howard Welser, Thomas Lento, and Marc Smith
	Electric Power Systems Restructuring Reliability and Cyber Security Paper Title: Infrastructure Protection in the Ancient World Author: Michael Assante
	Electronic Government Minitrack: E-Government Organization and Management Co-chairs: Helmut Krcmar, J. Ramon Gil-Garcia, and Maddalena Sorrentino Paper Title: Understanding the “Boundary” in Information Sharing and Integration Authors: Lei Zheng, Tung-Mou Yang, Theresa Pardo, and Yuanfu Jiang
	Information Technology in Healthcare Minitrack:T Adoption and Evaluation in Healthcare Paper Title: Methodological Approaches to Measuring the Effects of Implementation of Health Information Technology (HIT) Authors: Lance Roberts, Marcia Ward, Jane Brokel, Douglas Wakefield, Donald Crandall, and Paul Conlon
	Internet and the Digital Economy Minitrack: Social Networks and Virtual Worlds for Work, Learning and Play Paper Title: Cyber Migration: An Empirical Investigation on Factors that Affect Users’ Switch Intentions in Social Networking Sites Authors:Zengyan Cheng, Yinping Yang, and John Lim
	Knowledge Management Systems Minitrack: Knowledge Management Tools, Technologies, Mashups, and Design Approaches Co-chairs: Stefan Smolnik, Nassim Belbaly, and Richard Orwig Paper Title: The Visual Wiki: A New Metaphor for Knowledge Access and Management Authors: Christian Hirsch, John Hosking, John Grundy, Tim Chaffe, David MacDonald, and Yuriy Halytskyy
	Organizational Systems and Technology Social Issues in Information Technology Paper Title: Making it Hard to Lie: Cultural Determinants of Media Choice for Deception Authors: Christopher Furner and Joey George
	Software Technology Minitrack: Architectures for Distributed Systems Paper Title: Celebrating Diversity in Volunteer Computing Authors: David Anderson and Kevin Reed

They conclude that communities are a powerful way for businesses to grow.  Community and social media have ROI!

The segment is getting crowded, over 100 vendors and growing, but only few cleared Forrester’s threshold for robustness and feature richness.

Forrester evaluated nine social media platforms in depth and called out Jive Software and Telligent Systems on the strengths of the administrative and platform features and the company’s support and customization track record.

The other leaders included in the report: KickApps and Pluck, Awareness, Lithium Technologies, Mzinga, LiveWorld and Leverage Software.

This is great news for Telligent, of course, and we are happy to provide a copy of the Forrester report just for asking (and trading some contact information): http://telligent.com/forrester-report/.

Others are noting that social analytics are key differentiators in social media platforms:

http://tomhumbarger.wordpress.com/2009/01/15/ive-seen-the-social-analytics-future/

Telligent’s Harvest Social Media reporting platform provides a rich set of features for tracking activity in your community.

Knowledge sharing and Yahoo Answers: Everyone knows something

In particular, this image (figure 4) is a great use of an innovative way of handling large network graphs: chop them into a matrix of ego-net thumbnails.

Adamic et al. WWW 2008 Yahoo Answers Roles and Tag Ecologies

This is a neat way to side step the “blob” problem of many directed graph visualizations: too many nodes and too many links make the image impossible to understand.

Each grid of images represents a collection of authors who share contribution to questions with the same tag.  In this case the programming, wrestling, and marriage tags.  Each grid is a collection of ego-centric network diagrams, each author is displayed with their “1.5 degree” connections: their links to friends and thei friend’s links to one another.

Display a collection of authors and contrast multiple collections and  several interesting observations are possible.  First, not everyone who contributes to a tag is the same, a few highly active people make significant contributions while most people are lightly connected and make modest contributions.  Second, not everyone who contributes heavily does so in the same way.   In the upper left hand corner of each grid is the “top” person in that sample of the tag population.  Each is highly active but create different types of patterns of connection through that activity.  In programming the top person is a classic “answer person” – high out degree, low in-degree, connected to isolates and a resulting low clustering coefficient.  The top contributor in the Marriage tag is different, however: most of the people they connect to are connected to the other people they connect to: their “friends are friends”.  Their clustering coefficient is comparatively high in contrast with the top contributor to the “Programming” tag.  The top contributor of questions with the “Wrestling” tag is a hybrid: the author maintains a cluster of highly inter-connected repeat discussion partners while replying to a population of question people like a classic “answer person”.

It is worth noting the marriage resembles wrestling more than programing.

It is also worth noting that this visualization approach, while not perfect, is a nice step forward for information visualizations of complex graphs.  Graph vizualization has been stuck for many years: complex graphs are hard to draw in meaningful ways, let alone to do so automatically. This approach side-steps many of the obstacles to the main approach of whole graph visualization to focus on attributes of individuals and distributions of network variation.

The NodeXL add-in for Excel can generate these sub-graphs for any network: select “Insert subgraph images”.  The thumbnail of each node’s “egonet” is inserted in the spreadsheet and can be written to a local directory and later stitched into an array.

NodeXL Subgraph images