You Are Who You Talk To: Detecting Roles in Usenet Newsgroups

Fisher, Danyel; Smith, Marc A.; Welser, Howard T.

doi:10.1109/hicss.2006.536

Cited by 200 publications

(153 citation statements)

References 8 publications

Supporting

Mentioning

144

Contrasting

Unclassified

Order By: Relevance

“…The topic of the message is a joint function of the agent's interests, topics of the messages the agent has recently viewed, and the topic of the replied-to message if it is a reply. Theory and empirical evidence (Fisher, Smith & Welser, 2006 ;Faraj & Johnson, 2011 ) suggest three common patterns of interaction among community members: (1) preferential attachment, in which members respond to popular messages or posters; (2) reciprocity, in which members respond to others who have written to them in the past; and (3) interest matching, in which members respond to messages that match their interests. We thus assumed that agents in the model choose to reply to a message based on the average of (1) the number of replies the message has received; (2) the number of times the poster of the message has responded to the agent; and (3) the match between message topic and the agent's interests.…”

Section: Following the Roadmap: Using Abm To Inform The Design Of Onlmentioning

confidence: 99%

Agent Based Modeling to Inform the Design of Multiuser Systems

Ren

Kraut

2014

Ways of Knowing in HCI

View full text Add to dashboard Cite

Decades have passed since the inception of the fi eld of Human-Computer Interaction. The emergence of the Internet has shifted researchers' attention from understanding how individuals interact with computers to understanding how individuals interact with one another using computer technologies. A wide range of systems designed for multiple users have been labeled as groupware, collaborative computing, multiuser applications, and more recently social computing technologies. Designing these types of system is more challenging than designing singleuser ones because other people and their behaviors are integral elements of the system as experienced by users (Grudin, 1994 ). The system itself, therefore, is nondeterministic and evolutionary because the experience of some users is partly the result of decisions that earlier users have made. Because the behavior of a multiuser system is not stable until a critical mass of users has developed a routine way of using it, it is diffi cult to predict how groups of users will respond to a particular design before the stable state is reached. As a result, interactive design and evaluation cycle, perhaps the most successful HCI technique for system design, is insuffi cient for the design of multiuser applications.Consider the design of an online health support group like breastcancer.org and the decision about whether to employ moderators who ensure group members spend their time discussing cancer-related topics, channel off-topic content to sub-forums, or prevent users from posting advertisements. A member's decision to participate in

show abstract

Section: Following the Roadmap: Using Abm To Inform The Design Of Onlmentioning

confidence: 99%

Agent Based Modeling to Inform the Design of Multiuser Systems

Ren

Kraut

2014

Ways of Knowing in HCI

View full text Add to dashboard Cite

show abstract

“…In that case, the degree centrality is computed by the influence domain of the vertex. For a non directed graph G := (V, E, E), the influence domain of a vertex V i is the number or proportion of all other vertices which are connected by a path to that particular vertex [4]- [6].…”

Section: Trust Measurementmentioning

confidence: 99%

“…Our assumption is that credible users issue relevant answers. Researchers in various fields [4]- [6] have applied Social Network Analysis (SNA) method to different types of networks. To measure Influence value, in this paper, we analyze characteristics of Q&A networks by using SNA.…”

Section: Introductionmentioning

confidence: 99%

“…For measuring trust, we use the analysis of degree centrality which is frequently used when measuring authority and influence value [4]- [6]. Considering a graph G := (V, E A ) and E A the set of the directed connections E = (e 1,1 , e 1,2 , · · · , e i, j ) between the users of the set V then the in-degree centrality, K in i , and out-degree centrality, K out i of a vertex V i are the sum of the inbound connections and outbound connections to that vertex, respectively [4]- [6].…”

Section: Trust Measurementmentioning

confidence: 99%

See 1 more Smart Citation

InfluenceRank: Trust-Based Influencers Identification Using Social Network Analysis in Q&A Sites

Park

Seo

Lee

et al. 2012

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYQuestion and Answering (Q&A) sites are recently gaining popularity on the Web. People using such sites are like a communityanyone can ask, anyone can answer, and everyone can share, since all of the questions and answers are public and searchable immediately. This mechanism can reduce the time and effort to find the most relevant answer. Unfortunately, the users suffer from answer quality problem due to several reasons including limited knowledge about the question domain, bad intentions (e.g. spam, making fun of others), limited time to prepare good answers, etc. In order to identify the credible users to help people find relevant answer, in this paper, we propose a ranking algorithm, InfluenceRank, which is basis of analyzing relationship in terms of users' activities and their mutual trusts. Our experimental studies show that the proposed algorithm significantly outperforms the baseline algorithms. key words: knowledge sharing, question answering, social network analysis, identifying influencers

show abstract

“…According to the activity of the members and their patterns of communication, we defined several patterns based on the analyis of literature [8,13,32,48] like questioners, answering persons, trolls, conversationalists, and spammers. A questioner is a member, who seeks help and information.…”

Section: Retrieval and Exploration Toolsmentioning

confidence: 99%

Pattern-Based Cross Media Social Network Analysis for Technology Enhanced Learning in Europe

Klamma

Spaniol

Cao

et al. 2006

Innovative Approaches for Learning and Knowledge Sharing

View full text Add to dashboard Cite

Abstract. It is extremely challenging to get an overview of the current state-of-the-art in technology enhanced learning in Europe. Rapid technological and pedagogical innovations, constantly changing markets, a vivid number of small and medium enterprises, complex policy processes, ongoing political and societal debates on the pros and cons of technology enhanced leaning, combined with many languages and different cultures, make it almost impossible for people to be informed. We want to introduce the media base and the measure tools for pattern-based cross media social network analysis, created by the PROLEARN network of excellence in professional learning. The main goal of this endeavour is the reduction of complexity for actors in digital social networks by applying ideas from social software and already successful methods for complexity reduction, such as information visualization, social network analysis and pattern languages.

show abstract

You Are Who You Talk To: Detecting Roles in Usenet Newsgroups

Abstract: Abstract

Cited by 200 publications

References 8 publications

Agent Based Modeling to Inform the Design of Multiuser Systems

Agent Based Modeling to Inform the Design of Multiuser Systems

InfluenceRank: Trust-Based Influencers Identification Using Social Network Analysis in Q&A Sites

Pattern-Based Cross Media Social Network Analysis for Technology Enhanced Learning in Europe

Contact Info

Product

Resources

About