Feedback effects between similarity and social influence in online communities

Published on Aug 24, 2008 in KDD (Knowledge Discovery and Data Mining)
· DOI :10.1145/1401890.1401914
David J. Crandall31
Estimated H-index: 31
(Cornell University),
Dan Cosley35
Estimated H-index: 35
(Cornell University)
+ 2 AuthorsSiddharth Suri25
Estimated H-index: 25
(Cornell University)
A fundamental open question in the analysis of social networks is to understand the interplay between similarity and social ties. People are similar to their neighbors in a social network for two distinct reasons: first, they grow to resemble their current friends due to social influence; and second, they tend to form new links to others who are already like them, a process often termed selection by sociologists. While both factors are present in everyday social processes, they are in tension: social influence can push systems toward uniformity of behavior, while selection can lead to fragmentation. As such, it is important to understand the relative effects of these forces, and this has been a challenge due to the difficulty of isolating and quantifying them in real settings. We develop techniques for identifying and modeling the interactions between social influence and selection, using data from online communities where both social interaction and changes in behavior over time can be measured. We find clear feedback effects between the two factors, with rising similarity between two individuals serving, in aggregate, as an indicator of future interaction -- but with similarity then continuing to increase steadily, although at a slower rate, for long periods after initial interactions. We also consider the relative value of similarity and social influence in modeling future behavior. For instance, to predict the activities that an individual is likely to do next, is it more useful to know the current activities of their friends, or of the people most similar to them?
Figures & Tables
  • References (24)
  • Citations (479)
📖 Papers frequently viewed together
1,243 Citations
2001KDD: Knowledge Discovery and Data Mining
2,004 Citations
2002KDD: Knowledge Discovery and Data Mining
1,262 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
#1Robin Pemantle (UPenn: University of Pennsylvania)H-Index: 33
The models surveyed include generalized Polya urns, reinforced random walks, interacting urn models, and continuous reinforced processes. Emphasis is on methods and results, with sketches provided of some proofs. Applications are discussed in statistics, biology, economics and a number of other areas.
353 CitationsSource
#1Dan Cosley (Cornell University)H-Index: 35
#2Dan Frankowski (UMN: University of Minnesota)H-Index: 17
Last. John Riedl (UMN: University of Minnesota)H-Index: 60
view all 4 authors...
Member-maintained communities ask their users to perform tasks the community needs. From Slashdot, to IMDb, to Wikipedia, groups with diverse interests create community-maintained artifacts of lasting value (CALV) that support the group's main purpose and provide value to others. Said communities don't help members find work to do, or do so without regard to individual preferences, such as Slashdot assigning meta-moderation randomly. Yet social science theory suggests that reducing the cost and ...
192 CitationsSource
#1Fernanda B. Viégas (IBM)H-Index: 35
#2Martin Wattenberg (IBM)H-Index: 40
Last. F. van Ham (IBM)H-Index: 15
view all 4 authors...
Wikipedia, the online encyclopedia, has attracted attention both because of its popularity and its unconventional policy of letting anyone on the Internet edit its articles. This paper describes the results of an empirical analysis of Wikipedia and discusses ways in which the Wikipedia community has evolved as it has grown. We contrast our findings with an earlier study and present three main results. First, the community maintains a strong resilience to malicious editing, despite tremendous gro...
281 CitationsSource
#1Petter Holme (UM: University of Michigan)H-Index: 41
#2Mark Newman (UM: University of Michigan)H-Index: 91
Models of the convergence of opinion in social systems have been the subject of considerable recent attention in the physics literature. These models divide into two classes, those in which individuals form their beliefs based on the opinions of their neighbors in a social network of personal acquaintances, and those in which, conversely, network connections form between individuals of similar beliefs. While both of these processes can give rise to realistic levels of agreement between acquainta...
332 CitationsSource
#1Marcelo Cataldo (CMU: Carnegie Mellon University)H-Index: 20
#2Patrick Wagstrom (CMU: Carnegie Mellon University)H-Index: 11
Last. Kathleen M. Carley (CMU: Carnegie Mellon University)H-Index: 64
view all 4 authors...
Task dependencies drive the need to coordinate work activities. We describe a technique for using automatically generated archi-val data to compute coordination requirements, i.e., who must coordinate with whom to get the work done. Analysis of data from a large software development project revealed that coordina-tion requirements were highly volatile, and frequently extended beyond team boundaries. Congruence between coordination re-quirements and coordination activities shortened development t...
345 CitationsSource
Aug 20, 2006 in KDD (Knowledge Discovery and Data Mining)
#1Lars Backstrom (Cornell University)H-Index: 26
#2Daniel P. Huttenlocher (Cornell University)H-Index: 55
Last. Xiangyang Lan (Cornell University)H-Index: 5
view all 4 authors...
The processes by which communities come together, attract new members, and develop over time is a central research issue in the social sciences - political movements, professional organizations, and religious denominations all provide fundamental examples of such communities. In the digital domain, on-line groups are becoming increasingly prominent due to the growth of community and social networking sites such as MySpace and LiveJournal. However, the challenge of collecting and analyzing large-...
1,480 CitationsSource
Aug 20, 2006 in KDD (Knowledge Discovery and Data Mining)
#1Ravi Kumar (Yahoo!)H-Index: 73
#2Jasmine Novak (Yahoo!)H-Index: 14
Last. Andrew Tomkins (Yahoo!)H-Index: 41
view all 3 authors...
In this paper, we consider the evolution of structure within large online social networks. We present a series of measurements of two such networks, together comprising in excess of five million people and ten million friendship links, annotated with metadata capturing the time of every event in the life of the network. Our measurements expose a surprising segmentation of these networks into three regions: singletons who do not participate in the network; isolated communities which overwhelmingl...
978 CitationsSource
Social networks evolve over time, driven by the shared activities and affiliations of their members, by similarity of individuals' attributes, and by the closure of short network cycles. We analyzed a dynamic social network comprising 43,553 students, faculty, and staff at a large university, in which interactions between individuals are inferred from time-stamped e-mail headers recorded over one academic year and are matched with affiliations and attributes. We found that network evolution is d...
1,148 CitationsSource
#1Marshall Van Alstyne (MIT: Massachusetts Institute of Technology)H-Index: 28
#2Erik Brynjolfsson (MIT: Massachusetts Institute of Technology)H-Index: 63
Information technology can link geographically separated people and help them locate interesting or useful resources. These attributes have the potential to bridge gaps and unite communities. Paradoxically, they also have the potential to fragment interaction and divide groups. Advances in technology can make it easier for people to spend more time on special interests and to screen out unwanted contact. Geographic boundaries can thus be supplanted by boundaries on other dimensions. This paper f...
181 CitationsSource
Jan 23, 2005 in SODA (Symposium on Discrete Algorithms)
#1Noam Berger (California Institute of Technology)H-Index: 16
#2Christian Borgs (Microsoft)H-Index: 44
Last. Amin Saberi (Georgia Institute of Technology)H-Index: 44
view all 4 authors...
We analyze the contact process on random graphs generated according to the preferential attachment scheme as a model for the spread of viruses in the Internet. We show that any virus with a positive rate of spread from a node to its neighbors has a non-vanishing chance of becoming epidemic. Quantitatively, we discover an interesting dichotomy: for a virus with effective spread rate λ, if the infection starts at a typical vertex, then it develops into an epidemic with probability λΘ(log(1/λ)/log ...
163 Citations
Cited By479
#1Antonio Caliò (University of Calabria)H-Index: 2
#2Andrea Tagarelli (University of Calabria)H-Index: 20
Last. Francesco Bonchi (Institute for Scientific Interchange)H-Index: 42
view all 3 authors...
#1Wei Zhao (Xidian University)H-Index: 6
#2Ziyu Guan (Xidian University)H-Index: 16
Last. Xiaofei He (ZJU: Zhejiang University)H-Index: 56
view all 7 authors...
Social influence analysis in microblogging services has attracted much attention in recent years. However, most previous studies were focused on measuring users' (topical) influence. Little effort has been made to discern and quantify how a user is influenced. Specifically, the fact that user iretweets a tweet from author jcould be either because iis influenced by j(i.e., jis a topical authority), or simply because he is “influenced” by the content (interested in the content). To ...
#1Lingfei Li (Hangzhou Dianzi University)
#2Yezheng Liu (Hefei University of Technology)H-Index: 9
Last. Jiahang Yuan (Hangzhou Dianzi University)
view all 5 authors...
Abstract Information propagation modeling and influence maximization are two important research problems in viral marketing. When marketing information is given, how can the seed nodes be efficiently identified to maximize the spread of the information through the network? To answer this question, we consider multiple factors in information propagation, such as information content, social influence and user authority, and propose a multifactor-based information propagation model (MFIP). Then, we...
#1Jiayin Lin (Information Technology University)H-Index: 1
#2Geng Sun (Information Technology University)H-Index: 7
Last. Shiping Chen (CSIRO: Commonwealth Scientific and Industrial Research Organisation)H-Index: 22
view all 10 authors...
The soaring development of Web technologies and mobile devices has blurred time-space boundaries of people’s daily activities. Such development together with the life-long learning requirement give birth to a new learning style, micro learning. Micro learning aims to effectively utilize learners’ fragmented time to carry out personalized learning activities through online education resources. The whole workflow of a micro learning system can be separated into three processing stages: micro learn...
1 CitationsSource
#1Yukuo Cen (THU: Tsinghua University)H-Index: 1
#2Jing Zhang (RUC: Renmin University of China)H-Index: 2
Last. Jie Tang (THU: Tsinghua University)H-Index: 52
view all 8 authors...
This paper introduces how to infer trust relationships from billion-scale networked data to benefit Alibaba E-Commerce business. To effectively leverage the network correlations between labeled and unlabeled relationships to predict trust, we formalize trust into multiple types and propose a graphical model to incorporate type-based dyadic and triadic correlations, namely e-Trust. We also present a fast learning algorithm in order to handle billion-scale networks. we propose an efficient learnin...
1 CitationsSource
#1Luca LuceriH-Index: 5
#2Silvia GiordanoH-Index: 33
Last. Emilio Ferrara (SC: University of Southern California)H-Index: 38
view all 3 authors...
Since the 2016 US Presidential election, social media abuse has been eliciting massive concern in the academic community and beyond. Preventing and limiting the malicious activity of users, such as trolls and bots, in their manipulation campaigns is of paramount importance for the integrity of democracy, public health, and more. However, the automated detection of troll accounts is an open challenge. In this work, we propose an approach based on Inverse Reinforcement Learning (IRL) to capture tr...
1 Citations
#1Daniel Karell (NYUAD: New York University Abu Dhabi)H-Index: 2
#2Michael Freedman (HUJI: Hebrew University of Jerusalem)H-Index: 2
Abstract How do sociocultural dynamics shape conflict? We develop a relational understanding of how social relations, culture, and conflict are interwoven. Using this framework, we examine how combatants’ associations with cultural elements affect the interpersonal relationships underlying conflict dynamics, as well as how these relationships engender associations to cultural elements. To do so, we first introduce a novel analytical approach that synthesizes computational textual analysis and st...
1 CitationsSource
The second generation of World Wide Web, supplied with the platforms of Social Networks, proves to be a revealing source of knowledge. It has led to the emergence of online communities of interest. In fact, the interactions of users through web usages allow the exchange of information and the dissemination of innovation; thereby the formation of cohesive groups of individuals sharing goals, interests, semantics and services. Reflection on the evolution of those groups and their detection is a fu...
1 CitationsSource
#1Ruidong Yan (RUC: Renmin University of China)H-Index: 4
#2Yuqing Zhu (CSU: California State University)H-Index: 10
Last. Yongcai Wang (RUC: Renmin University of China)H-Index: 12
view all 4 authors...
Abstract Different from previous social influence problems such as Influence Maximization (IM), we in this paper first propose the Acceptance Probability Maximization (APM) problem, i.e., we select a seed set S with a budget b such that the acceptance probability of the target user set T is maximized. Then we employ the classical Independent Cascade (IC) model as the information diffusion model. Based on the IC model, we prove that the APM problem is NP-hard and the objective function is monoton...
#1Edwin N. Torres (UCF: University of Central Florida)H-Index: 15
ABSTRACTOnline communities are changing the way people enjoy leisure activities through computer-mediated interactions, yet little is known about how these groups form, function, and network. The current study draws on two years of involvement with a Meetup group, using participant observation and ethnographic interviews of the group's live interactions and content analysis of the group's online communication. Throughout this article, the author analyzed the life cycle of online communities and ...
4 CitationsSource