Download - Blogosphere: Research Issues, Tools and Applications · Web 2.0 Services (examples) • Blogs – Blogspot – Wordpress • Wikis – Wikipedia – Wikiversity • Social Networking

  • Blogosphere: Research Issues, Tools and Applications

    Huan Liu and Nitin Agarwal{Huan.Liu, Nitin.Agarwal.2}

    Computer Science and EngineeringArizona State University

    An updated version could be downloaded or��

  • Acknowledgments

    • We would like to express our sincere thanks to Magdiel Oliveras Galan, John J. Salerno, Shankar Subramanya, Sanjay Sundarajan,Lei Tang, Philip S. Yu , and Alan Zheng Zhao for collaboration, discussion, and valuable comments.

    • This work is, in part, sponsored by AFOSR and ONR grants in 2008.

    • This agreement covers the use of all slides of this tutorial.– You may use these slides freely for teaching if you send us an email

    stating the university name and class/course number in advance, and cite this tutorial.

    – If you wish to use these slides in any other ways, please contact (or email) us. The ppt version contains notes with additional information such as various sources in addition to References.

  • Outline

    • Background: Web 2.0 and Social Networks• Blogosphere: Definition, Types, and Comparison• Blogosphere Research Issues• Tools and APIs• Data Collection• Measures, Models, and Methods• Performance, Evaluation, and Metrics• Case Studies• References


  • Web vs. Web 2.0

  • Characteristics of Web 2.0

    • Rich Internet Applications• User generated contents• User enriched contents• User developed widgets• Collaborative environment: Participatory Web, Citizen


    • Thus, it leverages the power of the Long Tail with user generated data as the driving force

    • More of a paradigm shift than a technology shift

  • Web 2.0 Services (examples)• Blogs

    – Blogspot– Wordpress

    • Wikis– Wikipedia– Wikiversity

    • Social Networking Sites– Facebook– Myspace– Orkut

    • Digital media sharing websites– Youtube– Flickr

    • Social Tagging–

    • Others– Twitter– Yelp

  • Top 20 Most Visited Websites• Internet traffic report by Alexa on July 29th 2008

    • 40% of the top 20 websites are Web 2.0 sites

    1 Yahoo! 11 Orkut

    2 Google 12 RapidShare

    3 YouTube 13

    4 Windows Live 14 Microsoft Corporation

    5 Microsoft Network 15 Google India

    6 Myspace 16 Google Germany

    7 Wikipedia 17 QQ.Com

    8 Facebook 18 EBay

    9 Blogger 19 Hi5

    10 Yahoo! Japan 20 Google France

  • Social Networks

    • A social structure made of nodes (individuals or organizations) that are related to each other by various interdependencies like friendship, kinship, like, ...

    • Graphical representation– Nodes = members– Edges = relationships

  • Social Networks

  • Social Networks

    • A social structure made of nodes (individuals or organizations) that are related to each other by various interdependencies like friendship, kinship, like, ...

    • Graphical representation– Nodes = members– Edges = relationships

    • Various realizations– Social bookmarking (– Friendship networks (facebook, myspace)– Blogosphere – Media Sharing (Flickr, Youtube)– Folksonomies

  • • ACM TKDD Special Issue on Social Computing

    • Second International Conference on Social Computing, Behavioral Modeling, and Prediction (SBP09)

    • SIAM International Conf on Data Mining (SDM)Sparks (Reno area), Nevada, April 30 - May 2, 2009.

    Some Related CFPs���

  • BLOGOSPHEREDefinitions, Types, and Comparison

  • Blogging Phenomenon• It’s growing fast as a new means for online

    communications and interactions

    • A blogger could gain instant fame via his blogs

    • A blogger may make a good living with her blogs

    • Abundant, lucrative business opportunities• A new political arena

  • Source: The New York Times

    “The site, chock full of advertising, is a moneymaking machine – so much so that Ms. Armstrong and her husband have both quit their regular jobs.“The reason? The advertisers are eager to influence her 850,000 readers.

    Arnold Kim, founder and senior editor of

    “The site places MacRumors No. 2 on a list of the ‘25 most valuable blogs,’ …” What is the potential value? “Two of the other tech-oriented blogs on its list, …, were sold earlier this year, reportedly for sums in excess of $25 million.”

  • Blogosphere Growth• “In January 2004, there were about 1 million blogs on the

    Internet. As of mid-2006, the population of the ‘blogosphere’ was well past 50 million and climbing.” – Paul Gillin, The New Influencers, 2007

    “36 million women participate in the blogosphere each week, and 15 million have their own blogs”

    – A Study by BlogHer

    Today Front Page NY TimesThe Year of the Political Blogger Has Arrived… both parties understand the need to have greater numbers of bloggers attend.… to bring down the walls of the convention …

  • Understanding Blogosphere

    • Blogosphere• Blog sites• Bloggers• Blog posts• Reverse chronologically

    ordered entries • Blogroll• Permalinks• Trackback

    • Everyone can publish, but few are heard

    • Many interesting questions to address– How to build traffic– How to find niche online– How to increase

    influence– How to …

    • Fertile research domain

  • Blog Site

  • Blog Post

  • Blogger

  • Types of Blogs• Individual vs. community

    – Single authored (Individual blog sites)– Multi authored (Community blog sites)

    • Regulated vs. anonymous

    Individual Blog Sites Community Blog Sites

    Owned and maintained by individual users.Owned and maintained by a group of like-minded users.

    More like personal accounts, journals or diaries.More like discussion forums and discussion boards.

    No or almost negligible group interaction.High degree of group discussion and collaboration.

    No or almost negligible collective wisdom.Enormous collective wisdom and open source intelligence.

  • Blogosphere

    • Complex Social Networks• Vertices (Nodes): Bloggers/

    Blog posts/Blog sites

    • Edges: Relationships/Links• In-Degree: Number of


    • Out-Degree: Number of outlinks

  • Friendship Networks vs. BlogosphereFriendship Networks Blogosphere

    Explicit Links/Edges Implicit Links/Edges

    Undirected Graph Directed Graph

    Network Centrality Measures Blog Statistics

    Quantifying Spread of Influence Quantifying Influential Members

    Nodes are members/actors Nodes can be bloggers/blogs or blog sites

    Strictly defined graph structure Loosely defined graph structure

    “Being in touch” or “Making Friends” Sharing ideas and opinions

    Person-to-person Person-to-group

    Friendship Oriented Community Oriented

    Member’s Reputation/Trust based on network connections and/or location in the network

    Member’s Reputation/Trust based on the response to other member’s knowledge solicitations

  • Friendship Networks vs. Blogosphere

    Social Friendship Networks


    Social Networks

    Orkut, Facebook, LinkedIn,, etc.

    LiveJournal, MySpace, etc.

    TUAW, Blogger, Windows Live Spaces, etc.

  • Citation Networks vs. Blogosphere

    • Citation links – DBLP: strict notion of links. People cite what they refer to– Blogs: links are casual and often missing

    • Social networks– DBLP: inferred from co-authorship, citation networks– Blogs: people explicitly specify their social network or inferred

    from links, comments, etc.

    • Communities– DBLP: conference venues, journals, (relatively static)– Blogs: community blogs, inferred from blog roll (related blogs),

    topic taxonomy, blog-blog interaction, (very dynamic)


  • Understanding Blogosphere• Understand structures and properties of Blogosphere• Gain insights into the relationships between

    bloggers, readers, blog posts, comments, different blog sites in Blogosphere

    • Models help generate artificial data, tune the parameters to simulate special scenarios, and compare various studies and different algorithms

    • Study peculiarities in Blogosphere and infer latent patterns and structures that could explain certain phenomena like influence, diffusion, splogs, community discovery.

  • Modeling Web and Blogosphere• Some key differences between Web and Blogosphere

    – Models developed for Web assume dense graph structure due to a large number of interconnecting hyperlinks within webpages. This assumption does not hold true. Blogosphere is shown to have a very sparse hyperlink structure [Kritikopoulos et al. 2006].

    – The level of interaction in terms of comments and replies to blog posts makes Blogosphere different from Web

    – The highly dynamic and “short-lived” nature of the blog posts could not be simulated by the web models. Web models do not consider dynamicity in the web pages

    – Web models assume webpages accumulate links over time. However, this is not true with Blogosphere

    – “Categories” and “tags” gives blogs flexibility that conventional websites typically don’t have

    – Descriptive filenames used in permalinks of blogs as compared to webpage filenames

  • Modeling Blogosphere• Preferential attachment

    – Probability of a new edge to a node to be added depends on its degree– “The rich get richer”– Power law distribution or scale free distribution

    )deg():( iji vvveP ∝⇔

  • Modeling Blogosphere• Preferential attachment

    – Probability of a new edge to a node to be added depends on its degree– “The rich get richer”– Power law distribution or scale free distribution

    )deg():( iji vvveP ∝⇔

  • Modeling Blogosphere• Preferential attachment

    – Probability of a new edge to a node to be added depends on its degree– “The rich get richer”– Power law distribution or scale free distribution

    • Hybrid model– Mixture of both preferential attachment model and random model– Give a lucky poor guy some chance to get rich– To solve irreducibility (strong connectedness with few isolated subgraphs) random walk

    on a graph model proposes a random jump with a fixed probability

    • Leskovec et al. 2007 studied temporal patterns– How often people create blog posts– Busrtiness and popularity– How these posts are linked and what is the link density– Developed a SIS based model

    • Kumar et al. 2003 use blogrolls on the blog posts to construct a network of blog posts assuming that blogrolls contain similar blog posts

    VvvveP iji /)deg():( ∝⇔

    βαα )1(/)deg():( −+∝⇔ VvvveP iji

  • Blog Clustering

  • Blog Clustering• Dynamic and automatic organization of the content• Convenient accessibility• Optimizing search engines by reducing search space

    – Search only the relevant cluster• Focused crawling• Summarization• Topic identification• Reduce information overload

    – 175,000 blog posts per day, i.e., 2 blog posts per second – Dec 2006

    • Extraction and analysis of the trends

  • Blog Clustering (2)• Brooks and Montanez 2006, used tf-idf and

    picked top 3 keywords for blog posts– Clustered blogs based on these keywords– Reported improved clustering as compared to that using tags

    • Li et al. 2007 assigned different weights to title, body, and comments of blog posts – Need to address high dimensionality and sparsity due to their

    keyword-based approach

    • Agarwal et al. 2008 proposed a collective-wisdom based approach– Generate a category relation graph based on user assignments– Compute similarity matrix from this graph


    k jk

    jiji n




    { }jiji dtdD




    ijiji idftftfidf ⋅= ,,

  • Blog Mining• Interactions between producers and consumers improved with blogs• Consumers not only speak their mind but also broadcast their opinions• Blogs are invaluable information sources

    – consumers’ beliefs and opinions,– initial reaction to a launch,– understand consumer language, – track trends and buzzwords, and – fine-tune information needs

    • Blog conversations leave behind the trails of links, useful for understanding how information flows and how opinions are shaped and influenced

    • Tracking blogs also help in gaining deeper insights

  • Blog Mining for Opinion• A prototype system called Pulse [Gamon et al. 2005] uses a Naive Bayes

    classifier trained on manually annotated sentences with positive/negative sentiments and iterates until all unlabeled data is adequately classified.

    • Another system presented in [Attardi and Simi 2006] improves the blog retrieval by using opinionated words acquired from WordNet in the query proximity.

    • Some well-known opinion mining and sentiment analysis techniques [B. Liu 2006] could also be borrowed from text mining domain due to high textual nature of blogs.

    • LingPipe ( is another open source software which performs sentiment analysis on text corpora.

    – Subjective (opinion) vs. Objective (fact) sentences– Positive (favorable) vs. Negative (unfavorable) movie reviews

  • Influence

    • Market Movers: “word-of-mouth”, trust and reputation

    • Sway opinions: Government policies, campaign• Customer Support and Troubleshooting• Market research surveys: “use-the-views”• Representative articles: 18.6 new blog posts per sec• Advertising

  • Blog Influence• Two types of influence

    – Influential blog sites and site networks [Gill 2004, Gruhl et al 2004, Java et al 2006]

    – Influential bloggers in a community [Agarwal et al. 2008]

    • Blogosphere vs. Friendship Networks– Implicit vs. Explicit links– Blog statistics vs. Centrality measures– “influencing” vs. “could influence”– Loosely vs. Strictly defined graph structures

    • Blog vs. Webpage Ranking– Blog sites too sparse for webpage ranking algorithms to work [Kritikopoulos et

    al 2006]– Webpage acquires authority over time, blog posts’ influence diminishes– Greedy approach works better than PageRank, HITS to maximize influence

    flow [Kempe et al 2003, Richardson & Domingos 2002]

  • Issue of Trust• Open standards and low barriers to publishing have created

    overwhelming amount of collective wisdom• Yet more difficult for readers to discern whom to trust in

    some cases• Similar to WWW

    – Authoritative webpages e.g., HITS [Kleinberg et al. 1998], PageRank [Page et al. 1999]

    • Blogosphere allow mass to create and edit content compromising the sanctity of the original content

    • Some work exists for social friendship network domain, not many researchers have explored Blogosphere

    • Huge potential for trust study in Blogosphere domain

  • Trust• Kale et al. 2007 transformed the problem of trust in

    blogosphere to the one in social friendship networks– Studied propagation of trust among different blog sites– Mined sentiments from a window of words around hyperlinks– Identified positive, negative, or neutral sentiments towards the linked

    blog site

    – Constructed a network of blog sites using hyperlinks– Used Gruhl et al. 2004 trust propagation algorithm– Some concerns

    • These blog sites have to be linked for trust propagation• Trust is computed between blog sites based on how much one blog

    agrees or disagrees with the other

    Mi+1 = Mi * Ci – Perform till convergence

    M = Belief Matrix; Ci = Atomic Propagation

    Ci = M + MT*M + MT + M*MT

  • Community Extraction• Blogosphere doesn’t have an explicit notion of communities

    except community blogs

    • Discovering communities among individual blogs based on interaction

    • Different from blog clustering– Blog Clustering uses textual similarity– Community extraction taps interaction

    and link analysis

  • Community Extraction• Blogosphere doesn’t have an explicit notion of communities• Different from blog clustering• Researchers identify communities based on

    – Links: network of hyperlinks allows identification of virtual communities• Several studies on finding community of webpages like Kleinberg 1998

    and Kumar et al. 1999

    • While Kleinberg used authority and hubs idea to explore communities of webpages, Kumar et al. extended the idea of hubs and authorities and included co-citations as a way to extract all communities on the web and used graph theoretic algorithms to identify all instances of graph structures that reflect community characteristics.

    – Content: blogs with similar content or inspired by the same event form a virtual community

    • Kumar et al. 2003, Efimova and Hendrick 2005, Blanchard 2004

  • Community Extraction• Chin and Chignell 2006 proposed a model for finding

    communities taking the blogging behavior of bloggers into account– They aligned behavioral approaches through blog reader survey

    in studying blog community.

    • Blanchard and Marcus 2004 studied a multiple sport newsgroup “Virtual Settlement” and analyzed the possibility of emerging virtual communities– Newsgroups and discussion forums are similar in terms of

    interaction patterns to Blogosphere

    – More person-to-group interaction rather than person-to-person interaction

  • Spam blog (Splogs) Filtering• One of the major rising concerns on Blogosphere• Spammers make most of their money by getting viewers to click on ads that

    run adjacent to their nonsensical text

    • Open standards and low barriers to publishing escalates the problem and challenges while solving

    • Besides degrading search quality, affects the network resources

  • Spam blog (Splogs) Filtering• One of the major rising concerns on Blogosphere• Open standards and low barriers to publishing escalates the problem and

    challenges while solving

    • Besides degrading search quality, affects the network resources• Initial researches applied web spam link detection approaches

    – Ntoulas et al. 2006, distinguish between normal web pages and spam webpages based on the statistical properties like

    • number of words, average length of words, anchor text, title keyword frequency, tokenized URL

    – Gyongyi et al. 2004, Gyongyi et al. 2006 use PageRank to compute the spam score of a webpage

    • Kolari et al. 2006, consider each blog post as a static webpage and use both content and hyperlinks to classify a blog post as spam using a SVM based classifier

  • Spam blog (Splogs) Filtering• Some critical differences between web spam detection and

    splog detection – The content on blog sites is very dynamic as compared to that of web pages,

    so content based spam filters are ineffective

    – Moreover, spammers can copy the content from some regular blog posts to evade content based spam filters

    – Link based spam filters can easily be beaten by creating legitimate links

    • Lin et al. 2007, consider the temporal dynamics of blog posts and propose a self similarity based splog detection algorithm based on characteristic patterns found in splogs like, – Regularities or patterns in posting times of splogs, – Content similarity in splogs, and – Similar links in splogs.

  • Opinion and Sentiment Analysis• BLEWS (

    – Using Blogs to Provide Context for News Articles– Political views: Liberal vs. Conservative– Emotional charge

  • Opinion and Sentiment Analysis

  • Opinion and Sentiment Analysis

    • BLEWS (– Using Blogs to Provide Context for News Articles– Political views: Liberal vs. Conservative– Emotional charge

    • SKEWS (– Reveal bias in news story (articles)– Users rate the story on a scale from Liberal to Conservative– Readers vote

  • Opinion and Sentiment Analysis

  • Opinion and Sentiment Analysis• BLEWS (

    – Using Blogs to Provide Context for News Articles– Political views: Liberal vs. Conservative– Emotional charge

    • SKEWS (– Reveal bias in news story (articles)– Users rate the story on a scale from Liberal to Conservative– Readers vote

    • Opinion mining in legal blogs [Conrad and Schilder, 2007]– Collected blogs on legal search tools– N-gram Language modeling approach to determine

    • Subjectivity of text• Polarity of text• Degree of polarity


  • Analysis and Visualization Tools

    • Tools– Data Analysis & Visualization tools– Statistics like centrality measures

    • NetLogo (– Multi-agent programming language and modeling environment

    designed in Logo

    – Modelers can give instructions to hundreds or thousands of concurrently operating autonomous agents.

    – Exploring the connection between the individuals (micro-level) and the patterns that emerge from the interaction of many individuals (macro-level).

  • Analysis and Visualization Tools

    • StarLogo (– An extension of Logo– It is used to model the behavior of decentralized systems like social


    • REPAST (– Recursive Porous Agent Simulation Toolkit– Agent-based social network modeling toolkit. – It has libraries for genetic algorithms, neural networks, etc. and allows

    users to dynamically access and modify agents at run time.

    • Swarm ( Page)– A multi-agent simulation package– Simulates social or biological interaction of agents and their emergent

    collective behavior.

  • Analysis and Visualization Tools• UCINet (

    – Package for the analysis of social network data including centrality measures, subgroup identification, role analysis, elementary graph theory, and permutation-based statistical analysis

    – Has strong matrix analysis routines, such as matrix algebra and multivariate statistics

    • Pajek (– Slovenian for spider– Analyzing and visualizing large networks like social networks

    • Network package in R (– The network class can represent a range of relational data types, and

    support arbitrary vertex/edge/graph attributes– This is used to create and/or modify the network objects and is used

    for social network analysis (SNA)

  • Analysis and Visualization Tools

    • InFlow (– Integrated product for network analysis and visualization– Used in the SNA domain

    • NetMiner (– Tool for exploratory network data analysis and visualization– NetMiner allows to explore network data visually and

    interactively, and helps in detecting underlying patterns and structures of the network

  • APIs

    • APIs– Data collection (blog posts, inlinks, tags, etc.)– Technorati– Digg–– Facebook – StumbleUpon

  • Technorati API

    • bloginfo query API url:[apikey]&url=[blog url]

    Sample response:


    [blog name][blog URL][blog RSS URL][blog Atom URL][inbound blogs][inbound links][date blog last updated][blog ranking]

    [blog foaf URL]

  • Technorati API

    • BlogPostTags query API url:[apikey]&url=[blog url]

    Sample response:

    [limit parameter]

    [tag name];/tag>[tag count]

  • Digg API

    • List StoriesApi url:[epoch(07/01/2008)]&max_submit_date=[epoch(07/15/1008)]&appkey=[appkey]

    Sample response:

  • World's First Jailbroken iPhone 3G

    We can't say this is a surprise... but it is sweet to see. The iPhone Dev Team has added a video to their blog showing off the latest version of their upcoming PwnageTool 2.0, along with a video of what they claim is the "world's first" jailbroken iPhone 3G.

    Digg API

    lists 10 most popular stories from between 1st July 2008 and 15th July


  • API

    Returns a list of tags and number of times used

    Sample response


  • Some Available Datasets

    • Nielsen Buzzmetrics dataset (– ~ 14M blog posts from 3M blog sites collected by Nielsen BuzzMetrics

    in May 2006

    – 1.7M blog-blog links – Up to a half of the blog outlinks are missing– 51% of the total blog posts are in English

    • Enron Email dataset (– Emails from about 150 users– The corpus contains a total of about 0.5M messages– People have studied the social networks between users based on link


    – Links are constructed based on email senders and recipients

  • Available Datasets (2)

    • TREC (– A crawl of Feeds, and associated Permalink and homepage

    documents (from late 2005 and early 2006)

    – 100,649 feeds were polled once a week for 11 weeks– Total Number of Feeds collected:753,681 – Average feeds collected every day:10,615– Uncompressed Size:38.6GB Compressed Size:8.0GB– Reasonably sized spam component for added realism– Fee: £400 ~ $794.36

  • Available Datasets (3)

    • Mobile Network (– 27 objects – over 180,000 links – 1 object attribute – 2 link attributes

    • Other ways– Crawl blogs– Blogcatalog– Statistics available from technorati API– Tagging available from API

  • Data Crawler

    • BlogTrackers– User interface to crawl blog sites

    • Scratch crawling (from blog archives)• Incremental crawling (from RSS feeds)

    – Stores the blog posts in Microsoft SQL server– Collects

    – Track blog posts like generate tag clouds for user specified time window

    Blog post title Blog post tags

    Blog post content Blog post permalink

    Outlinks Blogger name

    Inlinks Blog post date and time

  • Collectable Statistics from Blogs

    • Inbound links– Blogs, blog post, webpage

    • Outbound links– Blogs, blog post, webpage

    • Comments• Blog server logs• Subscribers• Time to read/length• Links to post and incoming traffic from them• Links from post and outgoing traffic to them• Topic frequency score• Blogroll links• Tagged urls (, furl)

  • Citation Dataset• DBLP (

    – Over 1,200,000 objects– Over 2,480,000 links – 12 object attributes – 6 link attributes– 910 MB


  • Measures, Models, and Methods

    • Centrality Measures• Mathematical models: random, scale-free,

    preferential attachment, hybrid, cascade• Content analysis techniques• Link analysis• Supervised/unsupervised learning algorithms• Decision theoretic approaches• Agent-based modeling

  • Centrality Measures

    • Degree centrality– Defined as the number of ties a node has

    – For directed network• Indegree ~ “popularity”• Outdegree ~ “gregariousness”

    – O(V2) for V vertices in dense network– O(E) for E edges in sparse network

    },0),(:{)( jvvMevC jadjd ∀≠=

  • Centrality Measures• Betweenness centrality

    – a centrality measure of a vertex within a graph

    – Vertices that occur on many shortest paths between other vertices have higher betweenness than those that do not

    – Act as “broker” or “bridge”– O(V3) complexity– O(V2logV+VE) for sparse network




    tsVtvs st



    σ )()(

    σst is the geodesic path between s and t. σst(v) is the geodesic path between s and t passing through v.

  • Centrality Measures• Closeness centrality

    – A centrality measure of a vertex within a graph– Vertices that tend to have short geodesic distances to

    other vertices within the graph have higher closeness.

    – Defined as the mean geodesic distance between a vertex v and all other reachable vertices

    – O(V3) complexity1






  • Centrality Measures

    • Eigenvector centrality– Measure of the importance of a node in a network– Assigns relative scores to all nodes in the network– Better to connect to more “popular” nodes than

    less “popular” ones

    – Google's PageRank is a variant of the Eigenvector centrality measure



    jjjii xAx


    1λ or xAx



  • Mathematical Models

    • Power law– Polynomial relationship with scale invariance

    – a and α are constants > 1

    α−= axxf )(

    Power Law plot Log-log plot of Power Law

  • Mathematical Models

    • Power law– Examples: fractals, inverse square law, Zipf law,

    pareto rule, etc.– Two aspects of real networks (e.g., Social

    networks, Blog networks, World Wide Web, biological networks, etc.) make power law models an appropriate choice as compared to random models

    • Number of nodes (N) in the real networks is not static• Most real networks exhibit preferential connectivity.

  • Mathematical Models

    • Random– Random network models assume the probability that two vertices are

    connected is random and uniform

    • Preferential attachment– For example, a newly created webpage will be more likely to include links to

    well-known documents with already high connectivity

    – Thus the probability with which a new vertex connects to the existing vertices is not uniform

    – This property of power law models is also known as preferential attachment models

    • Hybrid– Pennock et al. 2002, have shown the relative importance of hybrid models in

    simulating social networks

    – Determine the appropriate proportion of random and scale free networks

    VvvveP iji /)deg():( ∝⇔

    βαα )1(/)deg():( −+∝⇔ VvvveP iji

    10,):( ≤≤∝⇔ ββji vveP

  • Mathematical Models• Cascade

    – Model information diffusion across the network– Linear threshold model

    • Assumes a linear relation between influencing and influenced nodes• Defines influencing capacity and tolerance limit of each node• Sum of the influencing capacities of the neighboring nodes > tolerance

    limit of this node, then this node gets influenced

    – Independent cascade model• Assumes the process of influence flow as cascade of events• Event represents a node being influenced• Each node is assigned an influencing probability• If node v influences node w then at time t+1 w gets influenced. No more

    attempts are made by v to influence w• Algorithm terminates when it is not possible to influence anymore nodes

  • Content Analysis Techniques

    • Blogs have rich textual content• Not only people create new content, they also enrich

    the existing content by providing meta data such as labels and tags

    • Human-generated tags are also called folksonomies• State-of-the-art content analysis techniques could be

    used for basic clustering, classification of the blog posts/blog sites

  • Content Analysis Techniques

    • tf-idf could be used for indexing the blog entries• Folksonomies could be considered as class labels• Supervised machine learning could be performed

    and learned models could be used to predict the tags of unlabeled corpus

    • This forms an essential concept for semi-automatically generating tag-clouds with least human intervention.

  • Link Analysis• Directed graph representation of blogs• Links form the edges of this graph

    – Incoming links (inlinks)– Outgoing links (outlinks)

    • Link analysis helps in understanding several interesting phenomena of social networks.

    • Text around the links give us knowledge about the linked blog posts.

    • Based on the links, hubs and authorities could be discovered. • This approach could lead to the identification of expert(s)

    within communities. • Link traversal: O(dh) for average outdegree d and h hops

  • Use of Link Analysis

    • Sparsity in the link structure of social networks makes it different from the World Wide Web model

    • Many of them like Blogosphere assume implicit link information among bloggers

    • Links could be constructed using the topic analysis

    • Blog posts talking about same topic could be connected– Supervised learning algorithms could be used to predict topics of

    unlabeled blogs

  • Decision Theoretic Approaches

    • Group-individual interaction and the effect of decision on an individual and/or a community as a whole.

    • Decision theory studies what is the best possible decision to take given a fully informed decision maker.

    • In social networks find the node that is the best to make decisions with least possible side-effects and maximum possible gains for the rest of the nodes. – Finding a node that has maximum information diffusion across

    • The analysis of such social decisions is dealt through game theory.

  • Agent-based Modeling• Each node in a social network can be treated as an agent

    [Sallach and Macal, 2001]

    • This agent could be a blogger in the blogosphere• Decision making ability of the agent can be modeled


    • This can help us in studying the factors that affect his/her blogging behavior, what and how (s)he makes decisions

    • Neural networks or genetic algorithms could also be used to train the model of these agents to closely simulate real-world scenario [Axelrod and Tesfatsion, 2005]


  • Performance• Does a project make any difference? We need to compare

    – Previously proposed model(s)– Baseline model(s)

    • Basic criteria– Efficiency (speed, scalability)– Correctness (get what you aim to get)

    • Traditional data mining/ machine learning performance criteria– Precision– Recall– F-measure– Area under ROC curve– Inter and intra cluster distances

    • Often we assume some ground truth• Training-testing models work on this assumption

    Train Test

    Total number of examples

  • Evaluation Challenges in Blogosphere

    • Concepts like influence, trust in Blogosphere can be subjective and often change based on particular needs

    • No ground truth available• Typical training-testing models may not work• Often resort to human evaluation and surveys

    – How to select subjects, and how many would suffice– How big is the evaluation budget, how long is the duration

    • Need to figure out objective ways of evaluation

  • Evaluation and Metrics• Obviously, various tasks may require different

    ways of performance evaluation– Blog search and retrieval– Clustering– Classification– Spam blogs– Diffusion– Influence

    • We provide some illustrative examples next.

  • Blog Search and Retrieval• Precision and Recall

    – Typically evaluated on unordered sets of documents

    – Top k results generate k sets for different values of k

    – P and R evaluated at different top k

    Recall Interpolated Precision

    0.0 1.00

    0.1 0.67

    0.2 0.63

    0.3 0.55

    0.4 0.45

    0.5 0.41

    0.6 0.36

    0.7 0.29

    0.8 0.13

    0.9 0.10

    1.0 0.08

    • Interpolated Precision– Defined as the highest precision at certain


    – Red line in the graph above shows the interpolated precision

    )(max)( rprprrip


  • Blog Search and Retrieval• Mean Average Precision (MAP)

    – Average of the precision scores after each relevant document retrieved for each query

    – Mean of the individual average precision scores for all the queries q є Q

    – Gives both precision and recall oriented aspects– Generates a single value for the set of queries– Less obvious interpretation than other measures

    ∑ ∑= =








    QMAP1 1


  • Measuring a Ranked List• Normalized Discounted Cumulative Gain (NDCG)

    • Measuring relevance of returned search result• Multi levels of relevance (r): irrelevant (0), borderline (1), relevant (2)• Each relevant document contributes some gain to be cumulated• Gain from low ranked documents is discounted• Normalized by the maximum DCG



    iin rddCG

    11 ),...,(




    in i

    rrddDCG2 2

    11 log),...,(






    2 21 log

    MaxDCGddDCGddNDCG nn /),...,(),...,( 11 =

  • NDCG - Example


    Ground Truth Ranking Function1 Ranking Function2

    Document Order



    Document Order


    1 d4 2 d3 2 d3 2

    2 d3 2 d4 2 d2 1

    3 d2 1 d2 1 d4 2

    4 d1 0 d1 0 d1 0

    NDCGGT=1.00 NDCGRF1=1.00 NDCGRF2=0.9203











    1 =






    2 =


    6309.4== GTDCGMaxDCG

    4 documents: d1, d2, d3, d4

  • Comparing Two Ranked Lists

    • Rank correlation– Spearman’s rank correlation


    – Exampleρ = 1-(6*194/10*(102-1))

    = -0.175


    1 22

    −−= ∑


    Xi Yirank


    rank yi

    di di2

    86 0 1 1 0 0

    97 20 2 6 -4 16

    99 28 3 8 -5 25

    100 27 4 7 -3 9

    101 50 5 10 -5 25

    103 29 6 9 -3 9

    106 7 7 3 4 16

    110 17 8 5 3 9

    112 6 9 2 7 49

    113 12 10 4 6 36

  • Concordance between a Pair• Rank correlation

    – [-1,1]: perfect agreement=1, perfect disagreement=-1– Kendall tau rank correlation coefficient

    – Example1




    Person A B C D E F G H

    Rank by Height 1 2 3 4 5 6 7 8

    Rank by Weight 3 4 1 2 5 7 8 6

    P = 5 + 4 + 5 + 4 + 3 + 1 + 0 + 0 = 22τ = (4*22/8*7 )-1= (88/56)-1 = 0.57

  • Blog Clustering• Within cluster between cluster distance

    – Small within cluster distance Cohesive – Large between cluster distance well-separated clusters

    • Distance between cluster mean/centroids• Single linkage• Complete linkage• Average linkage

    Cluster Mean/Centroids Single Linkage Complete Linkage Average Linkage

    Cohesive, well-separated clusters�

  • Blog Clustering• How many clusters should we have

    – The elbow criterion can be used to pick the number of clusters– Explained variance is ratio of between-group variance to total variance

  • Spam Blogs• Train-Test model• Precision, Recall, F-measure based metrics• Precision (P) = TP/(TP+FP)• Recall (R)= TP/(TP+FN)• F-measure (F) = 2*PR/(P+R)

    spam not-spam

    spam 7 4

    not-spam 3 6





    TP=7, FP=4, FN=4, TN=6P=7/11, R=7/10, F=0.663


    Where can we find FP, FN,TP, and TN


  • Case Studies

    • “Familiar Strangers” in Blogosphere• Employing Collective Wisdom• Blog Community Interaction• iFinder: Finding Influential Bloggers


  • Short Head and Long Tail

    • Few people are densely connected: Short Head

    • Many people are sparsely connected: Long Tail

    • Businesses like Amazon, Netflix, Wal-Mart, etc. obey this phenomenon

    • Wal-Mart sells more Long Tail items than Short Head

    • Zipf, Power Law, Pareto’s Law generate Long Tail

    Short Head

    Long Tail

  • Who are Familiar Strangers?

    • Observe repeatedly, but do not know each other• Real World– E.g., Individuals observe each other daily on a train– Discover the latent pattern: going to same workplace, • Blogosphere– What you write is what you are…– Have similar blogging behavior, interests (Movie and

    games, Technology, and Politics, etc.)

    – Never cited (came across) each other

  • Bloggers in Long Tail

    • Not returned as top hits by search engines• Not popular• Inordinately many• Disconnected• Movie Critics – Short Head


    • Movie Bloggers – Long Tail• Most lucrative test-bed for Familiar Strangers

  • Aggregating Niches in Long Tail

    • A blogger’s familiar-strangers together form a critical mass such that– the understanding of one blogger gives us a sensible

    and representative glimpse to others,– more data about familiar strangers can be collected

    for better customization and services (e.g., personalization and recommendation),

    – the nuances among them present new business opportunities, and

    – knowledge about them can facilitate predictive modeling and trend analysis.

  • Need for Aggregation• Customized attention requires

    substantial data• Majority of blog sites are in the

    Long Tail• …and are disconnected• Aggregating the similar yet

    disconnected for obtaining critical mass

    • Lack of data can result in irrelevant ads (see an example on the right)

    • Increase participation• Move from the Long Tail closer to

    the Short Head• Smooth knowledge transfer

    between familiar strangers

  • Definition – Familiar Strangers

    • Given a blogger b, familiar strangers to b are a set of bloggers B = {b1,b2,…,bn}, who share common patterns as b, like blogging on similar topics, but have never come across each other or have never related to each other.

    • Familiar:

    Blog posts

  • Definition – Familiar Strangers

    • Strangers:– Partial strangers– Total strangers

    • Partial strangers

    bj is in b’s Social Network b is in bj’s Social Network

  • Definition – Familiar Strangers

    • Total strangers

    • We focus on total strangers

    b and bj have disjoint Social Networks

  • Types of Familiar Strangers

    • Organizational differences in the blogosphere eventuate disparate types of familiar stranger bloggers

    Community-level familiar strangers

    Networking-site-level familiar strangers

    Blogosphere-level familiar strangers

  • Community Level Familiar Stranger

    • MySpace has a community called “A group for those who love history”

    • It has 38 members• two members, “Maria”

    and “John” – blog profusely on the

    similar topic,– but they are not in each

    other’s social network.

  • Networking Site Level Familiar Stranger

    • 2 groups on MySpace, – The Samurai (32 members)– The Japanese Sword (84

    members)– Marc, top blogger on “The

    Samurai” and Jeff, top blogger on “The Japanese Sword” discuss about Japanese martial arts.

    – Neither of them is in the other’s social network.

    – This implies, though being active locally and discussing on the same theme, the two bloggers are still strangers.

  • Blogosphere Level Familiar Stranger

    • 2 different social networking sites, MySpace and Orkut. – The Samurai (32 members)

    from MySpace– Samurai Sword (29 members)

    from Orkut– Top bloggers from the

    respective communities in MySpace and Orkut, Marc and Anant, respectively, share the blogging theme but they are not in each others’ social network.

    – The above example illustrates the existence of blogosphere-level familiar strangers.

  • Challenges

    • Link analysis• Defining Similarity• Data collection• Experiments• Evaluation & Validation• Current tools & technologies search the Short


  • Search via Blog Posts

    Search via Blogger’s Blog Post

  • Search via Context

    Search via Blogger’s context

  • Leveraging User Contributions



  • What is Collective Wisdom?

    • Shared knowledge arrived at by individuals and groups, used to solve problems

    • Group wisdom or Co-intelligence• Blog Clustering

    – User generated content as well as user enriched content– A prominent feature of social web– Several users tag and categorize their blogs– Collective wisdom emerges

  • Why Collective Wisdom?

    • Challenges with traditional approaches– High dimensionality– Sparsity– Do not leverage collective wisdom– Require number of clusters a priori– Similarity measure

  • Blog Categories

    Blog level Tags

    Blog Post level Tags

    5 Most recent blog posts’ snippets


  • BlogCatalog taxonomy

    WisClus clusters

  • Data Collection

    • Blogcatalog, using 4 bloggers as seed, crawled their social network in a breadth-first fashion

    • Report number of unique bloggers recorded with different number of seed bloggers (2,4,6)










    l Blo


    s Cr



    Total Number of Starting Bloggers

  • Dataset Characteristics

    • Variations in the dataset – depending on the category taxonomy– Top-level– All-category– One node-split: because of the skewed distribution of categories













    & e







    & m






















    & d







    & …




    e &





    og re














    ber o

    f Blo

    g Si


  • Experiments & Results

    • Link strength experiments: LinkStrength > 5• Category taxonomy variations: All-category• Baseline vs. WisClus

    – K-means– Hierarchical

    Type Method Within Avg Between Avg

    Baseline - BloggerSpaceKmeans 0.0363 0.2194Hierarchical 0.0890 0.3644

    WisClus - CategorySpaceKmeans 0.0615 0.2860Hierarchical 0.0857 0.2761

    WisClus - BloggerSpaceKmeans 0.0844 0.7090Hierarchical 0.0849 0.8118

  • Visualizations of clusters using Collective Wisdom

    Visualization Results

  • Visualizations of clusters using Baseline approach

    Visualization Results

  • Use Pajek to visualize the results

    Visualization Results


  • Blog Community Interaction Types

    • Discover community interaction through links�����������������

  • Interaction Through Observation

    • Interaction through observed events– Communities with similar sentiments could be aggregated


    Dislike Like







  • Proposed Approach – Flowchart

    Identify an event

    E.g., Saddam Hussein’s Death Sentence

    Analyze pre-event, during-event, post-event blog posts

    E.g., November-06, December-06, January-07

    Summarize the blog posts to pick relevant


    Generate Tag CloudsUse “WeFeelFine” API to filter the sentiments

    Compare these Sentiments to observe

    the interaction with respect to an event

  • A Running Example

    accept according agree

    Americaannounced Baghdad building cabinet decisions defense dialogue first future haveincrease looking mass partnerpatriotic people plan political powers regional see shares situation

    solutions start state term will

    army bad beginning channelscountry dead demonstrationsdown justice new occupation outside right Saddam Salahuddin security shut since single some stupidity todayZawra

    LegendPositive SentimentNegative Sentiment

    J F M …. … D J F M … … D J F M … N D J F …. … D J F M … D20082004 2005 2006 2007

    Iraq the Model Baghdad Burning

    Saddam’s Verdict


  • Physical and Virtual World

    Physical World

    Domain Expert


    Virtual World

    Online Community

  • Introduction

    • Inspired by the analogy between real-world and blog communities, we answer:

    Who are the influentials in Blogosphere?

    Can we find them?

    Active Bloggers = Influential Bloggers?

    • Active bloggers may not be influential• Influential bloggers may not be active

  • Searching The Influentials

    • Active bloggers– Easy to define– Often listed at a blog site– Are they necessarily influential

    • How to define an influential blogger?– Influential bloggers have influential posts– Subjective– Collectable statistics– How to use these statistics

  • Intuitive Properties• Social Gestures (statistics)

    – Recognition: Citations (incoming links)– An influential blog post is recognized by many. The more influential the

    referring posts are, the more influential the referred post becomes.– Activity Generation: Volume of discussion (comments)

    – Amount of discussion initiated by a blog post can be measured by the comments it receives. Large number of comments indicates that the blog post affects many such that they care to write comments, hence influential.

    – Novelty: Referring to (outgoing links)– Novel ideas exert more influence. Large number of outlinks suggests that

    the blog post refers to several other blog posts, hence less novel. – Eloquence: “goodness” of a blog post (length)

    – An influential is often eloquent. Given the informal nature of Blogosphere, there is no incentive for a blogger to write a lengthy piece that bores the readers. Hence, a long post often suggests some necessity of doing so.

    • Influence Score = f(Social Gestures)

  • A Preliminary Model• Additive models are good to determine the combined value of

    each alternative [Fensterer, 2007]. It also supports preferential independence of all the parameters involved in the final decision. A weighted additive function can be used to evaluate trade-offs between different objectives [Keeney and Raiffa, 1993].











    m nnoutmin








    −= ∑ ∑= =



    ι θ

  • Understanding the Influentials• Are influential bloggers simply active bloggers? • If not, in what ways are they different?

    – Can the model differentiate them? • Are there different types of influential bloggers?• What other parameters can we include to evolve

    the model?

    • Are there temporal patterns of the influential bloggers?

  • How to Evaluate the Model

    • Where to find the ground truth?– Lack of Training and Test data– Any alternative?

    • About the parameters – How can they be determined– Are they all necessary?

    • Are any of these correlated?• Data collection

    – A real-world blog site– “The Unofficial Apple Weblog”

  • Active & Influential Bloggers

    • Active and Influential Bloggers• Inactive but Influential Bloggers• Active but Non-influential Bloggers

    • We don’t consider “Inactive and Non-influential Bloggers”, because they seldom submit blog posts. Moreover, they do not influence others.

  • Lesion Study

    • To observe if any parameter is irrelevant.

  • Other Parameters

    • Rate of Comments

    “Spiky” comments reaction “Flat” comments reaction

  • Temporal Patterns of Influential Bloggers

    • Long term Influentials• Average term Influentials• Transient Influentials• Burgeoning Influentials

  • Verification of the Model

    • Revisit the challenges– No training and testing data– Absence of ground truth– Subjectivity

    • We use another Web 2.0 website, Digg as a reference point.

    • “Digg is all about user powered content. Everything is submitted and voted on by the Digg community. Share, discover, bookmark, and promote stuff that‘s important to you!”

    • The higher the digg score for a blog post is, the more it is liked.

    • A not-liked blog post will not be submitted thus will not appear in Digg.�

  • Verification of the Model• Digg records top 100 blog posts.

    • Top 5 influential and top 5 active bloggers were picked to construct 4 categories

    • For each of the 4 categories of bloggers, we collect top 20 blog posts from our model and compare them with Digg top 100.

    • Distribution of Digg top 100 and TUAW’s 535 blog posts

  • Verification of the Model• Observe how much our model aligns with Digg.

    • Compare top 20 blog posts from our model and Digg.

    • Considered last six months

    • Considered all configuration to study relative importance of each parameter.

    • Inlinks > Comments > Outlinks > Blog post length

  • Some Call for Papers• ACM TKDD Special Issue on Social Computing

    • Second International Conference on Social Computing, Behavioral Modeling, and Prediction (SBP09)

    • SIAM International Conf on Data Mining (SDM)Sparks (Reno area), Nevada, April 30 - May 2, 2009.���

  • References[Adar and Adamic, 2005] Adar, E. and Adamic, L. A. (2005). Tracking information epidemics in blogspace. In WI ’05: Proceedings of

    the The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI’05), pages 207–214, Washington, DC, USA. IEEE Computer Society.

    [Adar et al., 2004] Adar, E., Zhang, L., Adamic, L., and Lukose, R. (2004). Implicit structure and the dynamics of blogspace. InProceedings of the 13th International World Wide Web Conference.

    [Agarwal et al., 2007a] Agarwal, N., Galan, M., and Chen, Y. (2007a). Approximate structural matching over ordered xml documents. In International Database Engineering and Applicaion Symposium.

    [Agarwal et al., 2008a] Agarwal, N., Galan, M., Liu, H., and Subramanya, S. (2008a). Clustering blogs with collective wisdom. InProceedings of the International Conference on Web Engineering (ICWE08).

    [Agarwal et al., 2005] Agarwal, N., Haque, E., Liu, H., and Parsons, L. (2005). Research paper recommender system: A subspaceclustering approach. In The 6th International Conference on Web-Age Information Management (WAIM 2005), pages 475 –491.

    [Agarwal et al., 2006a] Agarwal, N., Haque, E., Liu, H., and Parsons, L. (2006a). A subspace clustering framework for research group collaboration. International Journal of Information Technology and Web Engineering, 1(1):35 – 58.

    [Agarwal and Liu, 2008a] Agarwal, N. and Liu, H. (2008a). Blogosphere: Research issues, tools, and appli-cations. SIGKDD Explorations.

    [Agarwal and Liu, 2008b] Agarwal, N. and Liu, H. (2008b). A study of communities and in fluence in blogosphere. In 2nd SIGMOD PhD Innovative Database and Research Doctorate Consortium (IDAR08), Vancouver, Canada.

    [Agarwal et al., 2008b] Agarwal, N., Liu, H., Salerno, J. J., and Sundarajan, S. (2008b). Understanding group interaction in blogosphere: A case study. In 2nd International Conference on Computational Cultural Dynamics (ICCCD08), Washington D.C.

    [Agarwal et al., 2007b] Agarwal, N., Liu, H., Salerno, J. J., and Yu, P. S. (2007b). Searching for Familiar Strangers on Blogosphere: Problems and Challenges. In NSF Symposium on Next-Generation Data Mining and Cyber-enabled Discovery and Innovation (NGDM).

  • References[Agarwal et al., 2008c] Agarwal, N., Liu, H., Tang, L., and Yu, P. S. (2008c). Identifying the in fluential bloggers. In Proccedings of the

    First ACM International Conference on Web Search and Data Mining (WSDM08) (Video available at: agarwal iib/).

    [Agarwal et al., 2006b] Agarwal, N., Liu, H., and Zhang, J. (2006b). Blocking objectionable web content by leveraging multiple information sources. SIGKDD Explor. Newsl., 8(1):17–26.

    [Albert, 2001] Albert, R. (2001). Statistical mechanics of complex networks. PhD thesis.

    [Aleman-Meza et al., 2006] Aleman-Meza, B., Nagarajan, M., Ramakrishnan, C., Ding, L., Kolari, P., Sheth,

    A. P., Arpinar, I. B., Joshi, A., and Finin, T. (2006). Semantic analytics on social networks: experiences in addressing the problem of conflict of interest detection. In WWW ’06: Proceedings of the 15th international conference on World Wide Web, pages 407–416, New York, NY, USA. ACM Press.

    [Ali-Hasan and Adamic, 2007] Ali-Hasan, N. and Adamic, L. (2007). Expressing social relationships on the blog through links and comments. In International Conference on Weblogs and Social Media.

    [Anderson, 2006] Anderson, C. (2006). The long tail : why the future of business is selling less of more. New York : Hyperion.

    [Attardi and Simi, 2006] Attardi, G. and Simi, M. (2006). Blog mining through opinionated words. In Proceedings of the fifteenth Text REtrieval Conference (TREC).

    [Avesani et al., 2005] Avesani, P., Massa, P., and Tiella, R. (2005). A trust-enhanced recommender system application: Moleskiing. In SAC, pages 1589–1593.

    [Backstrom et al., 2006] Backstrom, L., Huttenlocher, D., Kleinberg, J., and Lan, X. (2006). Group for-mation in large socialnetworks: membership, growth, and evolution. In KDD ’06: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 44–54, New York, NY, USA. ACM Press.

    [Barabasi and Albert, 1999] Barabasi, A. L. and Albert, R. (1999). Emergence of scaling in random net-works. Science, 286(509).

    [Bekkerman and McCallum, 2005] Bekkerman, R. and McCallum, A. (2005). Disambiguating web appear-ances of people in a social network. In WWW ’05: Proceedings of the 14th international conference on World Wide Web, pages 463–470, New York, NY, USA. ACM Press.

  • References[Blanchard and Markus, 2004] Blanchard, A. and Markus, M. (2004). The experienced sense of a virtual community:

    Characteristics and processes. The DATA BASE for Advances in Information Systems, 35(1).

    [Blum et al., 2006] Blum, A., Mugizi, T. H. C., and Rwebangira, M. R. (2006). A random-surfer web-graph model. In Third Workshopon Analytic Algorithmics and Combinatorics (ANALCO06).

    [Bonhard et al., 2006] Bonhard, P., Harries, C., McCarthy, J., and Sasse, M. A. (2006). Accounting for taste: using pro file similarity to improve recommender systems. In CHI ’06: Proceedings of the SIGCHI conference on Human Factors in computing systems, pages 1057–1066, New York, NY, USA. ACM Press.

    [Brin and Page, 1998] Brin, S. and Page, L. (1998). The anatomy of a large-scale hypertextual Web search engine. Computer Networks and ISDN Systems, 30(1–7):107–117.

    [Brooks and Montanez, 2006] Brooks, C. H. and Montanez, N. (2006). Improved annotation of the blogo-sphere via autotagging and hierarchical clustering. In WWW ’06: Proceedings of the 15th international conference on World Wide Web, pages 625–632, New York, NY, USA. ACM Press.

    [Cai et al., 2005] Cai, D., Shao, Z., He, X., Yan, X., and Han, J. (2005). Mining hidden community in heterogeneous social networks. In LinkKDD ’05: Proceedings of the 3rd international workshop on Link discovery, pages 58–65, New York, NY, USA. ACM Press.

    [Cai-Nicolas Ziegler, 2004] Cai-Nicolas Ziegler, G. L. (2004). Analyzing correlation between trust and user similarity in onlinecommunities. In iTrust, number 251-265.

    [Chi et al., 2006] Chi, Y., Tseng, B. L., and Tatemura, J. (2006). Eigen-trend: trend analysis in the blogo-sphere based on singular value decompositions. In CIKM ’06: Proceedings of the 15th ACM international conference on Information and knowledge management, pages 68–77, New York, NY, USA. ACM Press.

    [Chin and Chignell, 2006] Chin, A. and Chignell, M. (2006). A social hypertext model for finding community in blogs. In HYPERTEXT’06: Proceedings of the seventeenth conference on Hypertext and hypermedia, pages 11–22, New York, NY, USA. ACM Press.

  • References[Coffman and Marcus, 2004] Coffman, T. and Marcus, S. (2004). Dynamic classification of groups through social network analysis

    and hmms. In Proceedings of IEEE Aerospace Conference.

    [Conrad and Schilder, 2007] Conrad, J. G. and Schilder, F. (2007). Opinion mining in legal blogs. In ICAIL ’07: Proceedings of the 11th international conference on Arti ficial intelligence and law, pages 231–236, New York, NY, USA. ACM.

    [Deerwester et al., 1990] Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T. K., and Harshman, R. (1990). Indexing by latent semantic analysis. Journal of the American Society for information science.

    [Domingos and Richardson, 2001] Domingos, P. and Richardson, M. (2001). Mining the network value of customers. In KDD ’01: Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining, pages 57–66, New York, NY, USA. ACM Press.

    [Drezner and Farrell, 2004] Drezner, D. and Farrell, H. (2004). The power and politics of blogs. In American Political Science Association Annual Conference.

    [Durrett, 1988] Durrett, R. (1988). Lecture Notes on Particle Systems and Percolation. Wadsworth Pub-lishing.

    [Efimova and Hendrick, 2005] Efimova, L. and Hendrick, S. (2005). In search for a virtual settlement: An exploration of weblog community boundaries.

    [Elkin, ] Elkin, T. Just an online minute... online forecast. =Articles.showArticle art aid=29803.

    [Fensterer, 2007] Fensterer, G. D. (2007). Planning and Assessing Stability Operations: A Proposed Value Focus Thinking Approach. PhD thesis, Air Force Institute of Technology.

    [Flake et al., 2002] Flake, G., Lawrence, S., Giles, C. L., and Coetzee, F. (2002). Self-organization and identi fication of web communities. IEEE Computer, 35(3).

    [Flake et al., 2000] Flake, G. W., Lawrence, S., and Giles, C. L. (2000). E fficient identification of web communities. In 6th International Conference on Knowledge Discovery and Data Mining.

  • References[Friedman, 2005] Friedman, T. L. (2005). The World Is Flat: A Brief History of the Twenty-First Century. Farrar, Straus and Giroux.

    [Gamon et al., 2005] Gamon, M., Aue, A., Corston-Oliver, S., and Ringger, E. (2005). Pulse: Mining

    [Gamon et al., 2008] Gamon, M., et al. (2008). BLEWS: Using Blogs to Provide Context for News Articles. In 2nd ICWSM.

    Customer Opinions from Free Text. In Proceedings of the 6th International Symposium on Intelligent

    Data Analysis.

    [Gibson et al., 1998] Gibson, D., Kleinberg, J., and Raghavan, P. (1998). Inferring web communities from link topology. In 9th ACM Conference on Hypertext and Hypermedia.

    [Gill, 2004] Gill, K. E. (2004). How can we measure the influence of the blogosphere? In Proceedings of the WWW’04: workshop on the Weblogging Ecosystem: Aggregation, Analysis and Dynamics.

    [Gillmor, 2006] Gillmor, D. (2006). We the Media: Grassroots Journalism by the People, for the People. O’Reilly.

    [Girvan and Newman, 2002] Girvan, M. and Newman, M. E. J. (2002). Community structure in social and biological networks. In National Academy of Science.

    [Gladwell, 2000] Gladwell, M. (2000). The Tipping Point: How Little Things Can Make a Big Di fference. Little, Brown and Company.

    [Golbeck, 2006a] Golbeck, J. (2006a). Combining provenance with trust in social networks for semantic web content filtering. In IPAW, pages 101–108.

    [Golbeck, 2006b] Golbeck, J. (2006b). Generating predictive movie recommendations from trust in social networks. In iTrust, pages 93–104.

    [Golbeck et al., 2004] Golbeck, J., Bonatti, P. A., Nejdl, W., Olmedilla, D., and Winslett, M. (2004). Trust,

    security, and reputation on the semantic web. In Proceedings of the ISWC-04 Workshop on Trust, Security,

    and Reputation on the Semantic Web.

    [Golbeck and Hendler, 2004a] Golbeck, J. and Hendler, J. (2004a). Reputation Network Analysis for Email Filtering. In Proceedings Conference on Email and Anti-Spam (CEAS), Mountain View, USA.

  • References[Golbeck and Hendler, 2006] Golbeck, J. and Hendler, J. (2006). Inferring binary trust relationships in web-based social networks.

    ACM Trans. Inter. Tech., 6(4):497–529.

    [Golbeck and Hendler, 2004b] Golbeck, J. and Hendler, J. A. (2004b). Accuracy of metrics for inferring trust and reputation in semantic web-based social networks. In EKAW, pages 116–131.

    [Golbeck and Parsia, 2006] Golbeck, J. and Parsia, B. (2006). Trust network-based filtering of aggregated claims. IJMSO, 1(1):58–65.

    [Golbeck et al., 2003] Golbeck, J., Parsia, B., and Hendler, J. A. (2003). Trust networks on the semantic web. In CIA, pages 238–249.

    [Goldenberg et al., 2001] Goldenberg, J., Libai, B., and Muller, E. (2001). Talk of the network: A complex systems look at the underlying process of word-of-mouth. Marketing Letters, 12:211–223.

    [Gruhl et al., 2005] Gruhl, D., Guha, R., Kumar, R., Novak, J., and Tomkins, A. (2005). The predictive power of online chatter. In KDD ’05: Proceeding of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining, pages 78–87, New York, NY, USA. ACM Press.

    [Gruhl et al., 2004] Gruhl, D., Liben-Nowell, D., Guha, R., and Tomkins, A. (2004). Information di ffusion through blogspace. SIGKDD Exploration Newsletter, 6(2):43–52.

    [Guha et al., 2004] Guha, R., Kumar, R., Raghavan, P., and Tomkins, A. (2004). Propagation of trust and distrust. In WWW ’04:Proceedings of the 13th international conference on World Wide Web, pages 403–412, New York, NY, USA. ACM Press.

    [Gyongyi et al., 2006] Gyongyi, Z., Berkhin, P., Garcia-Molina, H., and Pedersen, J. (2006). Link spam detection based on mass estimation. In Proceedings of the 32nd International Conference on Very Large Data Bases (VLDB).

    [Gyongyi et al., 2004] Gyongyi, Z., Garcia-Molina, H., and Pedersen, J. (2004). Combating web spam with trustrank. In Proceedings of the 30th International Conference on Very Large Data Bases (VLDB).

    [Herring et al., 2005] Herring, S. C., Kouper, I., Paolillo, J. C., Scheidt, L. A., Tyworth, M., Welsch, P., Wright, E., and Yu, N. (2005). Conversations in the blogosphere: An analysis ”from the bottom up”. hicss, 04:107b.

  • References[Hopcroft et al., 2003] Hopcroft, J., Khan, O., Kulis, B., and Selman, B. (2003). Natural communities in large linked networks. In 9th

    Intl. Conf. on Knowledge Discovery and Data Mining. [Hope et al., 2006] Hope, T., Nishimura, T., and Takeda, H. (2006). An integrated method for social network extraction. In WWW

    ’06: Proceedings of the 15th international conference on World Wide Web, pages 845–846, New York, NY, USA. ACM Press. [Java et al., 2006] Java, A., Kolari, P., Finin, T., and Oates, T. (2006). Modeling the spread of in fluence on the blogosphere. In

    Proceedings of the 15th International World Wide Web Conference. [Kale et al., 2007] Kale, A., Karandikar, A., Kolari, P., Java, A., Finin, T., and Joshi, A. (2007). Modeling trust and in fluence in the

    blogosphere using link polarity. In International Conference on Weblogs and Social Media. [Katz and Golbeck, 2006] Katz, Y. and Golbeck, J. (2006). Social network-based trust in prioritized default logic. In AAAI. [Kautz et al., 1997] Kautz, H., Selman, B., and Shah, M. (1997). Referral web: combining social networks and collaborative filtering.

    Commun. ACM, 40(3):63–65. [Keeney and Raiffa, 1993] Keeney, R. L. and Raiffa, H. (1993). Decisions with Multiple Objectives: Preferences and Value Tradeoffs.

    Cambridge University Press. [Keller and Berry, 2003] Keller, E. and Berry, J. (2003). One American in ten tells the other nine how to vote, where to eat and,

    what to buy. They are The In fluentials. The Free Press. [Kleinberg, 1998] Kleinberg, J. (1998). Authoritative sources in a hyperlinked environment. In 9th ACM-SIAM Symposium on

    Discrete Algorithms. [Kolari et al., 2006a] Kolari, P., Finin, T., and Joshi, A. (2006a). SVMs for the blogosphere: Blog iden-ti fication and splog detection.

    In AAAI Spring Symposium on Computational Approaches to Analyzing Weblogs. [Kolari et al., 2006b] Kolari, P., Java, A., Finin, T., Oates, T., and Joshi, A. (2006b). Detecting spam blogs: A machine learning

    approach. In Proceedings of the 21st National Conference on Arti ficial Intelligence (AAAI). [Kritikopoulos et al., 2006] Kritikopoulos, A., Sideri, M., and Varlamis, I. (2006). Blogrank: ranking weblogs based on connectivity

    and similarity features. In AAA-IDEA ’06: Proceedings of the 2nd international workshop on Advanced architectures and algorithms for internet delivery and applications, page 8, New York, NY, USA. ACM Press.

  • References[Kumar et al., 2003] Kumar, R., Novak, J., Raghavan, P., and Tomkins, A. (2003). On the Bursty Evolution of Blogspace. In Proceedings of the

    12th international conference on World Wide Web, pages 568–576, New York, NY, USA. ACM Press. [Kumar et al., 2006] Kumar, R., Novak, J., and Tomkins, A. (2006). Structure and evolution of online social networks. In KDD ’06: Proceedings

    of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 611–617, New York, NY, USA. ACM Press.

    [Kumar et al., 1999] Kumar, R., Raghavan, P., Rajagopalan, S., and Tomkins, A. (1999). Trawling the web for emerging cyber communities. In The 8th International World Wide Web Conference.

    [Leshed and Kaye, 2006] Leshed, G. and Kaye, J. J. (2006). Understanding how bloggers feel: recognizing a ffect in blog posts. In CHI ’06: CHI ’06 extended abstracts on Human factors in computing systems, pages 1019–1024, New York, NY, USA. ACM Press.

    [Leskovec et al., 2007] Leskovec, J., McGlohon, M., Faloutsos, C., Glance, N., and Hurst, M. (2007). Cas-cading behavior in large blog graphs. In SIAM International Conference on Data Mining.

    [Li et al., 2007] Li, B., Xu, S., and Zhang, J. (2007). Enhancing clustering blog documents by utilizing author/reader comments. In ACM-SE 45: Proceedings of the 45th annual southeast regional conference, pages 94–99, New York, NY, USA. ACM Press.

    [Liggett, 1985] Liggett, T. (1985). Interacting Particle Systems. Springer. [Lin et al., 2006] Lin, Y.-R., Sundaram, H., Chi, Y., Tatemura, J., and Tseng, B. (2006). Discovery of blog communities based on mutual

    awareness. In Proceedings of the 3rd annual workshop on webloging ecosystem: aggreation, analysis and dynamics. [Lin et al., 2007] Lin, Y.-R., Sundaram, H., Chi, Y., Tatemura, J., and Tseng, B. L. (2007). Splog detection using self-similarity analysis on blog

    temporal dynamics. In Proceedings of the 3rd international workshop on Adversarial information retrieval on the web (AIRWeb), pages 1–8, New York, NY, USA. ACM Press.

    [Liu et al., 2007] Liu, Y., Huang, X., An, A., and Yu, X. (2007). Arsa: a sentiment-aware model for predicting sales performance using blogs. In SIGIR ’07: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval, pages 607–614, New York, NY, USA. ACM Press.

    [Marlow et al., 2006] Marlow, C., Naaman, M., Boyd, D., and Davis, M. (2006). Ht06, tagging paper, taxonomy, flickr, academic article, to read. In HYPERTEXT ’06: Proceedings of the seventeenth conference on Hypertext and hypermedia, pages 31–40, New York, NY, USA. ACM Press.

  • References[Massa and Avesani, 2004] Massa, P. and Avesani, P. (2004). Trust-aware collaborative filtering for recommender systems. In

    CoopIS/DOA/ODBASE, pages 492–508. [Massa and Avesani, 2005] Massa, P. and Avesani, P. (2005). Controversial users demand local trust metrics: An experimental

    study on community. In AAAI, pages 121–126. [Massa and Bhattacharjee, 2004] Massa, P. and Bhattacharjee, B. (2004). Using trust in recommender sys-tems: An experimental

    analysis. In iTrust, pages 221–235. [Massa and Hayes, 2005] Massa, P. and Hayes, C. (2005). Page-rerank: Using trusted links to re-rank authority. In Web Intelligence,

    pages 614–617. [Matsumura et al., 2005] Matsumura, N., Goldberg, D. E., and Llorà, X. (2005). Mining directed social network from message

    board. In WWW ’05: Special interest tracks and posters of the 14th inter-national conference on World Wide Web, pages 1092–1093, New York, NY, USA. ACM Press.

    [Matsuo et al., 2006] Matsuo, Y., Mori, J., Hamasaki, M., Ishida, K., Nishimura, T., Takeda, H., Hasida, K., and Ishizuka, M. (2006). Polyphonet: an advanced social network extraction system from the web. In WWW ’06: Proceedings of the 15th international conference on World Wide Web, pages 397–406, New York, NY, USA. ACM Press.

    [McDonald, 2003] McDonald, D. W. (2003). Recommending collaboration with social networks: a compar-ative evaluation. In CHI ’03: Proceedings of the SIGCHI conference on Human factors in computing systems, pages 593–600, New York, NY, USA. ACM Press.

    [McNee et al., 2002] McNee, S. M., Albert, I., Cosley, D., Gopalkrishnan, P., Lam, S. K., Rashid, A. M., Konstan, J. A., and Riedl, J. (2002). On the recommending of citations for research papers. In CSCW ’02: Proceedings of the 2002 ACM conference on Computer supported cooperative work, pages 116–125, New York, NY, USA. ACM Press.

    [Mei et al., 2007] Mei, Q., Ling, X., Wondra, M., Su, H., and Zhai, C. (2007). Topic sentiment mixture: modeling facets and opinions in weblogs. In WWW ’07: Proceedings of the 16th international conference on World Wide Web, pages 171–180, New York, NY, USA. ACM Press.

  • References[Mishne and de Rijke, 2006] Mishne, G. and de Rijke, M. (2006). Deriving wishlists from blogs show us your blog, and we’ll tell you

    what books to buy. In Proceedings of the 15th international conference on World Wide Web, pages 925–926, New York, NY, USA. ACM Press.

    [Newman, 2003] Newman, M. E. J. (2003). The structure and function of complex networks. SIAM Review, 45:167. [Newman, 2004b] Newman, M. E. J. (2004b). Fast algorithm for detecting community structure in networks. Physical Review E,

    69:066133. [Ntoulas et al., 2006] Ntoulas, A., Najork, M., Manasse, M., and Fetterly, D. (2006). Detecting spam web pages through content

    analysis. In Proceedings of the 15th international conference on World Wide Web (WWW). [O’Reilly, 2005] O’Reilly, T. (2005). What is Web 2.0 -design patterns and business models for the next generation of software. 2005/09/30/what-is-web-20.html. [Osman and Yearwood, 2007] Osman, D. J. and Yearwood, J. L. (2007). Opinion search in web logs. In ADC ’07: Proceedings of the

    eighteenth conference on Australasian database, pages 133–139, Darlinghurst, Australia, Australia. Australian Computer Society, Inc.

    [Page et al., 1998] Page, L., Brin, S., Motwani, R., and Winograd, T. (1998). The pagerank citation ranking: Bringing order to the web. Technical report, Stanford Digital Library Technologies Project.

    [Pennock et al., 2002] Pennock, D. M., Flake, G. W., Lawrence, S., Glover, E. J., and Giles, C. L. (2002). Winners don’t take all: Characterizing the competition for links on the web. Proceedings of the National Academy of Sciences, 99(8):5207–5211.

    [Pujol et al., 2002] Pujol, J. M., Sangesa, R., and Delgado, J. (2002). Extracting reputation in multi agent systems by means of social network topology. In Proceedings of the first international joint conference on Autonomous agents and multiagent systems (AAMAS), pages 467–474, New York, NY, USA. ACM Press.

    [Richardson and Domingos, 2002] Richardson, M. and Domingos, P. (2002). Mining knowledge-sharing sites for viral marketing. In Proceedings of the eighth ACM SIGKDD international conference on Knowledge Discovery and Data mining, pages 61–70, New York, NY, USA. ACM Press.

  • References[Sabater and Sierra, 2002] Sabater, J. and Sierra, C. (2002). Reputation and social network analysis in multi-agent systems. In

    AAMAS ’02: Proceedings of the first international joint conference on Autonomous agents and multiagent systems (AAMAS), pages 475–482, New York, NY, USA. ACM Press.

    [Schelling, 1978] Schelling, T. (1978). Micromotives and macrobehavior norton. [Scoble and Israel, 2006] Scoble, R. and Israel, S. (2006). Naked conversations : how blogs are changing the way businesses talk

    with customers. John Wiley. [Spertus et al., 2005] Spertus, E., Sahami, M., and Buyukkokten, O. (2005). Evaluating similarity measures: a large-scale study in

    the orkut social network. In Proceeding of the eleventh ACM SIGKDD international conference on Knowledge Discovery in Data mining (KDD), pages 678–684, New York, NY, USA. ACM Press.

    [Stefanone et al., 2004] Stefanone, M., Hancock, J., Gay, G., and Ingra ffea, A. (2004). Emergent networks, locus of control, and the pursuit of social capital. In Proceedings of the 2004 ACM conference on Computer Supported Cooperative Work (CSCW), pages 592–595, New York, NY, USA. ACM Press.

    [Tang et al., 2008] Tang, L., Liu, H., Zhang, J., Agarwal, N., and Salerno, J. J. (2008). Topic taxonomy adaptation for group profiling. ACM Transactions on Knowledge Discovery from Data, TKDD, 1(4).

    [Terveen and McDonald, 2005] Terveen, L. and McDonald, D. W. (2005). Social matching: A framework and research agenda. ACM Trans. Comput.-Hum. Interact., 12(3):401–434.

    [Thelwall, 2006] Thelwall, M. (2006). Bloggers under the London attacks: Top information sources and topics. In Proceedings of the 3rd annual workshop on webloging ecosystem: aggreation, analysis and dynamics.

    [Watts and Strogatz, 1998] Watts, D. J. and Strogatz, S. H. (1998). Collective dynamics of ‘small-world networks. Nature, 393(6684):440442.

    [Wu et al., 2003] Wu, F., Huberman, B. A., Adamic, L. A., and Tyler, J. (2003). Information flow in social groups. [Yu and Singh, 2003] Yu, B. and Singh, M. P. (2003). Detecting deception in reputation management. In Proceedings of the second international joint conference on Autonomous Agents and Multiagent Systems (AAMAS), pages 73–80,

    New York, NY, USA. ACM Press.

  • References[Zhang and Varadarajan, 2006] Zhang, Z. and Varadarajan, B. (2006). Utility scoring of product reviews. In Proceedings of the 15th

    ACM international conference on Information and Knowledge Management (CIKM), pages 51–57, New York, NY, USA. ACM Press.

    [Zhou and Davis, 2006] Zhou, Y. and Davis, J. (2006). Community discovery and analysis in blogspace. In Proceedings of the 15th international conference on World Wide Web, pages 1017–1018, New York, NY, USA. ACM Press.

    [Ziegler and Golbeck, 2007] Ziegler, C.-N. and Golbeck, J. (2007). Investigating interactions of trust and interest similarity. Decis. Support Syst., 43(2):460–475.

    [Ziegler and Lausen, 2004a] Ziegler, C.-N. and Lausen, G. (2004a). Paradigms for decentralized social filtering exploiting trustnetwork structure. In CoopIS/DOA/ODBASE (2), pages 840–858.

    [Ziegler and Lausen, 2004b] Ziegler, C.-N. and Lausen, G. (2004b). Spreading activation models for trust propagation. In Proceedings of the 2004 IEEE International Conference on e-Technology, e-Commerce and e-Service (EEE’04), pages 83–97, Washington, DC, USA. IEEE Computer Society.

    [Ziegler and Lausen, 2005] Ziegler, C.-N. and Lausen, G. (2005). Propagation models for trust and distrust in social networks. Information Systems Frontiers, 7(4-5):337–358.

    [Ziegler and Skubacz, 2006] Ziegler, C.-N. and Skubacz, M. (2006). Towards automated reputation and brand monitoring on the web. In WI ’06: Proceedings of the 2006 IEEE/WIC/ACM International Con-ference on Web Intelligence, pages 1066–1072, Washington, DC, USA. IEEE Computer Society.

    [Sallach and Macal, 2001] Sallach, David and Macal, Charles (2001). The simulation of social agents: an introduction, Special Issue of Social Science Computer Review 19(3):245–248.

    [Axelrod and Tesfatsion, 2005] Axelrod, Robert and Tesfatsion, Leigh (2005). A guide for newcomers to Agent-Based Modeling in the Social Sciences, Handbook of Computational Economics, Vol. 2: Agent-Based Computational Economics

    Blogosphere: Research Issues, Tools and ApplicationsAcknowledgmentsOutlineWEB 2.0 AND SOCIAL NETWORKSWeb vs. Web 2.0Characteristics of Web 2.0Web 2.0 Services (examples)Top 20 Most Visited WebsitesSocial NetworksSocial NetworksSocial NetworksA Little Detour�BLOGOSPHEREBlogging PhenomenonSlide Number 16Blogosphere GrowthUnderstanding BlogosphereSlide Number 19Slide Number 20Slide Number 21Types of BlogsBlogosphereFriendship Networks vs. BlogosphereFriendship Networks vs. BlogosphereCitation Networks vs. BlogosphereBLOGOSPHERE RESEARCH ISSUESUnderstanding BlogosphereModeling Web and BlogosphereModeling BlogosphereModeling BlogosphereModeling BlogosphereBlog ClusteringBlog ClusteringBlog Clustering (2)Blog MiningBlog Mining for OpinionInfluenceBlog InfluenceIssue of TrustTrustCommunity ExtractionCommunity ExtractionCommunity ExtractionSpam blog (Splogs) FilteringSpam blog (Splogs) FilteringSpam blog (Splogs) FilteringOpinion and Sentiment AnalysisOpinion and Sentiment AnalysisOpinion and Sentiment AnalysisOpinion and Sentiment AnalysisOpinion and Sentiment AnalysisTOOLS AND APISAnalysis and Visualization ToolsAnalysis and Visualization ToolsAnalysis and Visualization ToolsAnalysis and Visualization ToolsAPIsTechnorati APITechnorati APIDigg APIDigg APIDATA COLLECTIONSome Available DatasetsAvailable Datasets (2)Available Datasets (3)Data CrawlerCollectable Statistics from BlogsCitation DatasetMEASURES, MODELS, AND METHODSMeasures, Models, and MethodsCentrality MeasuresCentrality MeasuresCentrality MeasuresCentrality MeasuresMathematical ModelsMathematical ModelsMathematical ModelsMathematical ModelsContent Analysis TechniquesContent Analysis TechniquesLink AnalysisUse of Link AnalysisDecision Theoretic ApproachesAgent-based ModelingPERFORMANCE, EVALUATION, AND METRICSPerformanceEvaluation Challenges in BlogosphereEvaluation and MetricsBlog Search and RetrievalBlog Search and RetrievalMeasuring a Ranked ListNDCG - ExampleComparing Two