Feedback effects between similarity and social influence in online communities

By David J. Crandall, Dan Cosley, Dan Huttenlocher, Jon M. Kleinberg, Siddharth Suri. Knowledge Discovery and Data Mining, 2008

667 citations recorded by the source corpus.

Subfields: Communication, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/3d236be88d438bf21abd21ec415e2f2ac13e9aa1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Communication

In Statistical and Nonlinear Physics

Citations in the corpus

Friendship prediction and homophily in social media (2012, background): “Disentangling these scenarios is a delicate matter that requires longitudinal data sets, as social inﬂuence implies a temporal evolution of a relationship [Crandall et al. 2008; Aral et al. 2009; Aiello et al. 2010; Shalizi and Thomas 2010].”
Sequential Influence Models in Social Networks (2010, background): “Wikipedia has been the subject of research both because its entire history is freely available for analysis, and because, beyond serving as an encyclopedia, it is also a community (see e.g. (Viegas et al. 2007; Stvilia et al. 2005; Voss 2005; Wilkinson and Huberman 2007; Crandall et al. 2008)).”
A matrix factorization technique with trust propagation for recommendation in social networks (2010, background): “Ru,i can be any real number, but often ratings are integers in the range [1, 5].”
Modeling relationship strength in online social networks (2010, background): “[4] study the temporal evolution of link structure and attribute similarity in Wikipedia and propose a mathematical model that includes both influence and homophily effects to predict future behavior in the network.”
Finding social roles in Wikipedia (2011, background): “[9] observe a sharp rise in “similarity between two Wikipedia editors” (abstract category) as measured by overlap in edited pages (a behavioral regularity) at about the time they first communicate with each other (creating a network attribute) through a large-scale quantitiative analysis of editing ”
Distinguishing between Personal Preferences and Social Influence in Online Activity Feeds (2016, background): “In addition, people may be externally exposed to the same item through forces outside of the social network [19].”
Popularity versus similarity in growing networks (2011, background): “In similarity-based models, for example, such validation is difficult because there are too many different similarity measures, and it is usually unclear which one should be used in which case [22, 69], so that cases where model predictions are validated directly against real-world similarity data [”
Social Network Analysis and Mining for Business Applications (2011, methodology): “The information retrieval approach is exemplified by the P@NOPTIC Expert system described by Craswell et al. [2001]. In this system, first a collection of all the documents authored by an individual is collected; then, documents are concatenated to create a “person-document.”
Community Detection and Mining in Social Media (2010, background): “, 2006), and article edits in Wikipedia (Crandall et al., 2008), while some argue that there is no strong influence effect for the tagging behavior in Flickr (Anagnostopoulos et al.”
Social influence analysis in large-scale networks (2009, methodology): “[6] studies the correlation between social similarity and influence.”
Portrait of an Online Shopper: Understanding and Predicting Consumer Behavior (2015, background): “In reality, all these effects are interconnected [9, 3] and are difficult to disentangle from observational data [35].”
What makes an image popular? (2014, background): “They [4] suggest that users have an increased tendency to conform to their friends’ interests.”
The social distributional hypothesis: a pragmatic proxy for homophily in online social networks (2014, neutral): “(2) Crandall et al. (2008) discuss similarity and social influence in online communities, providing the general idea that friends interact similarly.”
Recommender Systems (2016, neutral): “The TrustWalker algorithm [269] is based on the observation that social network links provide an independent source of information from ratings [172].”