The role of social networks in information diffusion

Eytan BakshyItamar RosennCameron MarlowLada Adamic

article2012WWW1,663 citationsBest Paper Award

Proves through a massive randomized field experiment that weak ties drive the majority of novel information spread in online social networks, despite strong ties exerting greater individual influence.

Listen

Online social networks have transformed how individuals discover and distribute news and information. However, determining whether sharing behavior is truly caused by peer influence or merely reflects common interests and external media consumption (homophily and outside exposure) has long posed a major empirical challenge. Understanding the true causal mechanisms behind information diffusion is critical for organizations that rely on social channels for communication, outreach, and digital strategy.

The article evaluates the causal impact of social signals on online information diffusion and estimates the relative roles of close contacts ("strong ties") versus distant acquaintances ("weak ties") in spreading content. Using a massive, real-world randomized field experiment on Facebook involving 253 million users and over 1.1 billion user-link pairs over a seven-week period, the study randomized whether subjects were exposed to news feed stories about their friends' link-sharing activities or had those specific stories withheld.

The study yielded several key findings regarding how information spreads. First, exposure to social signals significantly increases sharing: subjects exposed to a link in their feed were approximately 7.37 times more likely to share it than those who were not exposed, and they shared it much faster (a median sharing latency of 6 hours compared to 20 hours). Second, while strong ties are individually more influential per interaction, weak ties collectively drive the vast majority of information dissemination across the network because they are far more numerous. Third, weak ties are the primary conduits for novel information, exposing users to content they would not have otherwise encountered, whereas exposure through strong ties exhibits substantial redundancy with external information sources. Finally, having multiple sharing friends increases a user's absolute likelihood of sharing, but the relative impact of feed exposure is highest when only a few friends share.

These findings indicate that digital information diffusion behaves predominantly as a "simple contagion" driven by the broad reach of weak ties, contrasting with complex behaviors that require dense, repeated reinforcement from close peers. For organizational decision-makers and digital strategists, relying exclusively on highly connected clusters or key influencers may underperform compared to strategies that leverage wide, decentralized networks of weak ties to introduce and circulate novel content. Decision-makers should evaluate their distribution and marketing campaigns by assessing both individual influence strength and structural network reach.

Because interactions occurring entirely outside the platform could not be observed, the findings reflect a lower bound on interpersonal influence and an upper bound on external correlation within the platform's ecosystem. Nevertheless, the study's randomized experimental design and unprecedented scale provide high confidence in the aggregate dynamics of online information diffusion.

Meta University of Michigan information diffusion information propagation information sharing information diffusion social networks social influence processes

arXiv: 1201.4145

Paper: Everyone's an influencer: quantifying influence on twitter, Eytan Bakshy et al. (2011). This paper establishes foundational empirical findings on information cascades and the limits of individual influence in online networks, directly setting up the source paper's experimental investigation into causal diffusion mechanisms.
Paper: Maximizing the spread of influence through a social network, David Kempe et al. (2003). This seminal study formalizes the fundamental algorithmic models of influence propagation and information cascade dynamics across social networks that the source paper evaluates experimentally in situ.
Paper: What is Twitter, a social network or a news media?, Haewoon Kwak et al. (2010). This work analyzes large-scale topology and information diffusion dynamics across social media networks, providing the structural context for the source paper's causal experiments on peer exposure.
Paper: Meme-tracking and the dynamics of the news cycle, J. Leskovec et al. (2009). This study models the temporal and structural dynamics of digital phrase propagation, establishing empirical baselines for tracking online information dissemination before the source's randomized trial.
Paper: Mining the network value of customers, Pedro M. Domingos et al. (2001). This work introduces the modeling framework for quantifying customer network value and social word-of-mouth propagation, underpinning subsequent empirical tests of network influence.

Paper: Design and analysis of experiments in networks: Reducing bias from interference, Dean Eckles et al. (2017). This paper directly addresses methodological challenges in estimating causal peer effects and network interference that arise in massive online field experiments like the one conducted in the source study.
Paper: Fake News Detection on Social Media: A Data Mining Perspective, Kai Shu et al. (2017). This survey applies principles of social network diffusion, echo chambers, and tie influence discovered in empirical studies like the source to the problem of detecting and curbing fake news.
Paper: The COVID-19 social media infodemic, Matteo Cinelli et al. (2020). This study applies real-world information diffusion and epidemiological transmission dynamics to track the cross-platform spread of public health narratives and misinformation.
Paper: Epidemic processes in complex networks, Romualdo Pastor-Satorras et al. (2015). This comprehensive review synthesizes mathematical models of epidemic and information spreading processes across complex, heterogeneous networks.
Paper: Graph Neural Networks for Social Recommendation, Wenqi Fan et al. (2019). This paper extends the study of social influence and varying tie strengths into deep learning architectures by modeling user-to-user social networks and interaction strengths.

Cover for The role of social networks in information diffusion

Abstract

Online social networking technologies enable individuals to simultaneously share information with any number of peers. Quantifying the causal effect of these technologies on the dissemination of information requires not only identification of who influences whom, but also of whether individuals would still propagate information in the absence of social signals about that information. We examine the role of social networks in online information diffusion with a large-scale field experiment that randomizes exposure to signals about friends' information sharing among 253 million subjects in situ. Those who are exposed are significantly more likely to spread information, and do so sooner than those who are not exposed. We further examine the relative role of strong and weak ties in information propagation. We show that, although stronger ties are individually more influential, it is the more abundant weak ties who are responsible for the propagation of novel information. This suggests that weak ties may play a more dominant role in the dissemination of information online than currently believed.

Topics

information diffusion information propagation information sharing online social networkings social relations user information network architecture

Keywords

information diffusion social networks social influence processes peer influence information contagion information sharing external influence communication networks media contagion

1 Introduction
2 Related Work
3 Experimental Design and Data
3.1 Assignment Procedure
3.2 Ensuring Data Quality
3.3 Population
3.4 Evaluating Outcomes
4 How Exposure to Social Signals Affects Diffusion
4.1 Temporal Clustering
4.2 Effect of Multiple Sharing Friends
5 Tie Strength and Influence
5.1 Effect of Tie Strength
5.2 Collective Impact of Ties
6 Discussion
7 Acknowledgments
References

Knowls

Knowl 1 — Randomized In Situ Feed Experiment for Identifying Social Influence
experimental setup
To isolate the causal effect of online social signals on information diffusion from homophily and external confounds, a large-scale randomized field experiment was conducted on Facebook over seven weeks (August 14 to October 4, 2010). The experiment evaluated sharing behavior across $253,238,367$ subjects, $75,888,466$ unique URLs, and $1,168,633,941$ subject-URL pairs.

When a user's friend shared a URL, the subject-URL pair was randomly assigned at the time of feed rendering to either a feed condition (where stories containing the link appeared normally in the subject's News Feed) or a no feed condition (where all stories containing that URL were suppressed from the subject's News Feed). The assignment was deterministic for a given subject-URL pair, ensuring all subsequent shares of the same URL by other friends remained in the same condition. To maximize statistical power, twice as many pairs were assigned to the no feed condition. Because removal operated on individual subject-URL pairs, shared URLs were still delivered to over $99\%$ of their potential audience across the network.

To ensure data quality and avoid pre-treatment exposure:
- Only URLs shared by friends after the start of the experiment were included.
- Subject-URL pairs were excluded if the subject had clicked the link anywhere on the site up to two months prior.
- If a subject interacted with a URL via any interface other than the feed (e.g., direct messages, wall posts), the pair was removed from the experiment.
- Malicious links and spam were filtered using site integrity classification systems.
Knowl 2 — Overall Causal Impact of Feed Exposure on Information Sharing
empirical result
Exposure to social signals in the Facebook News Feed substantially increases the likelihood that a user shares that information compared to unexposed peers.

Across the experimental population:
- The baseline probability of sharing a URL without feed exposure ( $p_{\text{no\ feed}}$ ) is $0.025\%$ .
- The probability of sharing a URL upon feed exposure ( $p_{\text{feed}}$ ) is $0.191\%$ .
- The resulting relative risk ratio is: $\frac{p_{\text{feed}}}{p_{\text{no\ feed}}} = 7.37 \quad (95\% \text{ CI: } [7.23, 7.72])$
- The average treatment effect on the treated (absolute risk increase) is $p_{\text{feed}} - p_{\text{no\ feed}} = 0.166\%$ .
Furthermore, conditional on clicking a URL received in the feed condition, on average $1$ out of every $12.5$ clicked URLs ( $8\%$ ) is subsequently re-shared by the user.
Knowl 3 — Collective Dominance of Weak Ties in Aggregate Information Diffusion
empirical result
While strong ties exert higher per-exposure influence on an individual, weak ties are responsible for the vast majority of information spread across the entire social network in aggregate.

When tie strength is partitioned into weak ties (zero direct interactions in the prior three months, $k=0$ ) and strong ties (at least one direct interaction, $k \ge 1$ ), the collective volume of diffusion driven by weak ties exceeds that driven by strong ties by a wide margin across all four operational definitions of tie strength (comments received, messages received, photo co-tags, and thread co-comments). The abundance of weak ties in users' News Feeds outweighs the higher individual transmission probability of strong ties.
Knowl 4 — Formulation of Collective Tie-Strength Influence
equation
The causal effect of exposure to information shared by friends of tie strength $k \in \mathbb{N}_0$ is defined as the average treatment effect on the treated: $\text{ATET}(k) = p(k, \text{feed}) - p(k, \text{no\ feed})$ where $p(k, \text{feed})$ and $p(k, \text{no\ feed})$ are the probabilities of sharing given exposure and non-exposure on the feed from a friend of tie strength $k$ , respectively.

Let $f(k)$ denote the fraction of all URL impressions in users' feeds that originate from friends of tie strength $k$ , such that $\sum_{k=0}^N f(k) = 1$ . Under a threshold where $k = 0$ designates weak ties and $k \ge 1$ designates strong ties, the total fractions of sharing events attributable to weak ties ( $T_{\text{weak}}$ ) and strong ties ( $T_{\text{strong}}$ ) are: $T_{\text{weak}} = \text{ATET}(0) \cdot f(0)$ $T_{\text{strong}} = \sum_{i=1}^N \text{ATET}(i) \cdot f(i)$ where $N$ is the maximum observed interaction frequency.
Knowl 5 — Individual Tie Strength, Influence, and Information Novelty
empirical result
For subjects exposed to exactly one friend sharing a URL, tie strength exhibits distinct relationships with individual influence, homophily, and information novelty:
1. Individual Influence: A subject's probability of sharing increases with tie strength in the feed condition. For example, subjects exposed to a link from a friend from whom they received three comments are $2.83$ times more likely to share than if they received zero comments.
2. Homophily and External Correlation: In the unexposed (no feed) condition, subjects are $3.84$ times more likely to share a link that was shared by a strong tie ( $3$ comments) than a weak tie ( $0$ comments), demonstrating that strong ties share substantially more common external information sources and interests.
3. Information Novelty (Relative Impact): The relative risk ratio $\frac{p(k, \text{feed})}{p(k, \text{no\ feed})}$ decreases as tie strength increases and is highest for weak ties ( $k=0$ ). This demonstrates that weak ties primarily transmit novel information that the user would not have otherwise encountered or shared.
Knowl 6 — Differential Absolute and Relative Effects of Multiple Sharing Friends
empirical result
As the number of friends who have shared a given Web link increases from $1$ to $6$ :
- The raw probability of sharing increases monotonically in both the feed condition (from $\approx 0.002$ to $\approx 0.021$ ) and the no feed condition (from $\approx 0.00025$ to $\approx 0.007$ ).
- The absolute causal effect of feed exposure ( $p_{\text{feed}} - p_{\text{no\ feed}}$ ) increases with the number of sharing friends (from $\approx 0.0017$ for $1$ friend to $\approx 0.014$ for $6$ friends), indicating that multiple social exposures have an increasing additive causal effect.
- The relative risk ratio ( $\frac{p_{\text{feed}}}{p_{\text{no\ feed}}}$ ) declines monotonically from $\approx 7.4$ for $1$ sharing friend to $\approx 3.0$ for $6$ sharing friends.
This divergence indicates that while social feed signals have the largest absolute impact when reinforced by multiple peers, content shared by multiple friends carries substantial redundancy with external information sources and homophily.
Knowl 7 — Interaction-Based Measures of Tie Strength
definition
Tie strength between a subject and a sharing friend is quantified using direct online and offline interaction counts logged over the three months immediately prior to the experiment, rather than static topological proxies such as mutual friend counts. Four independent metrics are defined:
1. Messages received: Frequency of private Facebook messages sent from the friend to the subject.
2. Comments received: Frequency of public comments posted by the friend on the subject's posts.
3. Photo coincidences: Number of distinct photographs on Facebook in which both users are co-tagged, capturing real-world shared events.
4. Thread coincidences: Number of distinct Facebook posts (by any user) on which both the subject and the friend posted comments.
A cutoff of $k = 0$ interactions defines a weak tie, while $k \ge 1$ interactions defines a strong tie.
Knowl 8 — Feed Acceleration of Information Diffusion and Baseline Temporal Clustering
empirical result
Exposure to information in the News Feed significantly accelerates the speed of information diffusion:
- For URLs assigned to both conditions, the median latency between a friend sharing a link and the subject sharing that same link is $6$ hours in the feed condition, compared to $20$ hours in the no feed condition (Wilcoxon rank-sum test, $p < 10^{-16}$ ).
- Users in the feed condition share predominantly immediately upon feed exposure.
- Strong temporal clustering occurs even in the no feed condition: subjects sharing without feed exposure still share shortly after their friends. This demonstrates that temporal clustering alone in observational network data is heavily confounded by shared external timing and common external media exposure.
Knowl 9 — Non-Feed Observational Bounds on Homophily and Influence
theoretical result
Because interactions taking place outside of the host platform (e.g., in-person conversations, email, external websites) cannot be directly observed, the experimental conditions provide theoretical bounds on the mechanisms driving on-site information sharing:
- Upper Bound on Homophily: The probability of sharing in the absence of feed exposure, $p(\text{no\ feed})$ , serves as an upper bound on how much sharing occurs purely due to homophily and shared external interests, as it also includes unobserved external peer influence (e.g., word-of-mouth or messaging outside the feed).
- Lower Bound on Social Influence: The absolute risk difference between conditions, $p(\text{feed}) - p(\text{no\ feed})$ , provides a conservative lower bound on the total amount of on-site sharing attributable to interpersonal influence across all communication channels.

Knowl 10 — Experimental Population Demographics and Balance

data/table

The experimental population comprised approximately $253$ million monthly active Facebook users across $236$ countries and territories (with $44$ countries having $\ge 1$ million subjects). The subject median age was $26$ (mean $29.3$ ). Subjects were assigned in a balanced fashion across the feed ( $N = 160,688,092$ ) and no feed ( $N = 218,743,932$ ) conditions:

Demographic Feature (% of subjects)	feed	no feed
Gender
Female	51.6%	51.4%
Male	46.7%	47.0%
Unspecified	1.5%	1.5%
Age
17 or younger	12.8%	13.1%
18–25	36.4%	36.1%
26–35	27.2%	26.9%
36–45	13.0%	12.9%
46 or older	10.6%	10.9%
Country (top 10 other)
United States	28.9%	29.1%
Turkey	6.1%	5.8%
Great Britain	5.1%	5.2%
Italy	4.2%	4.1%
France	3.8%	3.9%
Canada	3.7%	3.8%
Indonesia	3.7%	3.5%
Philippines	2.1%	2.3%
Germany	2.3%	2.3%
Mexico	2.0%	2.1%
226 Others	37.5%	37.7%

These balanced proportions demonstrate that the randomization across subject-URL pairs preserved covariate balance across key demographic segments.

Coverage note — No substantial contributed material was omitted.

References

1.L. A. Adamic and E. Adar. Friends and neighbors on the web. Social Networks, 25:211–230, 2001.
2.E. Adar and A. Adamic, Lada. Tracking information epidemics in blogspace. In 2005 IEEE/WIC/ACM International Conference on Web Intelligence, Compiegne University of Technology, France, 2005.
3.E. Adar, J. Teevan, and S. T. Dumais. Resonance on the web: web dynamics and revisitation patterns. In Proceedings of the 27th International Conference on Human factors in Computing Systems, CHI ’09, pages 1381–1390, New York, NY, USA, 2009. ACM Press.
4.A. Anagnostopoulos, R. Kumar, and M. Mahdian. Influence and correlation in social networks. In Proceedings of the 14th Internal Conference on Knowledge Discover & Data Mining, pages 7–15, New York, NY, USA, 2008. ACM Press.
5.S. Aral, L. Muchnik, and A. Sundararajan. Distinguishing influence-based contagion from homophily-driven diffusion in dynamic networks. Proc. Natl. Acad. Sci., 106(51):21544–21549, December 2009.
6.S. Aral and D. Walker. Creating social contagion through viral product design: A randomized trial of peer influence in networks. Management Science, 57(9):1623–1639, Aug. 2011.
7.L. Backstrom, D. Huttenlocher, J. Kleinberg, and X. Lan. Group formation in large social networks: 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, 2006. ACM.
8.E. Bakshy, J. M. Hofman, W. A. Mason, and D. J. Watts. Everyone’s an influencer: Quantifying influence on twitter. In 3rd ACM Conference on Web Search and Data Mining, Hong Kong, 2011. ACM Press.
9.E. Bakshy, B. Karrer, and L. Adamic. Social influence and the diffusion of user-created content. In Proceedings of the tenth ACM conference on Electronic commerce, pages 325–334. ACM, 2009.
10.H. R. Bernard, P. Killworth, D. Kronenfeld, and L. Sailer. The problem of informant accuracy: The validity of retrospective data. Annu. Rev. Anthropol., 13:495–517, 1984.
11.J. J. Brown and P. H. Reingen. Social ties and word-of-mouth referral behavior. J. Consumer Research, 14(3):pp. 350–362, 1987.
12.R. S. Burt. Structural holes: The social structure of competition. Harvard University Press, Cambridge, MA, 1992.
13.D. Centola. The Spread of Behavior in an Online Social Network Experiment. Science, 329(5996):1194–1197, September 2010.
14.D. Centola and M. Macy. Complex contagions and the weakness of long ties. Am. J. Sociol., 113(3):702–734, Nov. 2007.
15.M. Cha, A. Mislove, and K. P. Gummadi. A measurement-driven analysis of information propagation in the flickr social network. In Proceedings of the 18th international conference on World wide web, WWW ’09, pages 721–730, New York, NY, USA, 2009. ACM.
16.N. A. A. Christakis and J. H. H. Fowler. The spread of obesity in a large social network over 32 years. N. Engl. J. Med., 357(4):370–379, July 2007.
17.S. Fox. The social life of health information. Technical report, Pew Internet & American Life Project, 2011.
18.E. Gilbert and K. Karahalios. Predicting tie strength with social media. In Proceedings of the 27th International Conference on Human Factors in Computing Systems, CHI ’09, pages 211–220, New York, NY, USA, 2009. ACM.
19.M. Gladwell. The Tipping Point: How Little Things Can Make a Big Difference. Little Brown, New York, 2000.
20.M. Gomez Rodriguez, J. Leskovec, and A. Krause. Inferring networks of diffusion and influence. In Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining, KDD ’10, pages 1019–1028, New York, NY, USA, 2010. ACM.
21.A. Goyal, F. Bonchi, and L. V. Lakshmanan. Learning influence probabilities in social networks. In Proceedings of the third ACM international conference on Web search and data mining, WSDM ’10, pages 241–250, New York, NY, USA, 2010. ACM.
22.M. S. Granovetter. The strength of weak ties. Am. J. Sociol., 78(6):1360–1380, May 1973.
23.M. S. Granovetter. Threshold models of collective behavior. Am. J. Sociol., 83(6):1420–1443, 1978.
24.B. S. Greenberg. Person to person communication in the diffusion of news events. Journalism Quarterly, 41:489–494, 1964.
25.D. Gruhl, R. Guha, D. Liben-Nowell, and A. Tomkins. Information diffusion through blogspace. In Proceedings of the 13th international conference on World Wide Web, pages 491–501. ACM, 2004.
26.K. Hampton, L. S. Goulet, L. Rainie, and K. Purcell. Social networking sites and our lives. Technical report, Pew Internet & American Life Project, 2011.
27.S. Hill, F. Provost, and C. Volinsky. Network-Based marketing: Identifying likely adopters via consumer networks. Stat. Sci., 21(2):256–276, May 2006.
28.G. Kossinets and D. J. Watts. Origins of homophily in an evolving social network. Am. J. Sociol., 115(2):405–450, September 2009.
29.K. Lerman and R. Ghosh. Information contagion: An empirical study of the spread of news on digg and twitter social networks. In Proceedings of 4th International Conference on Weblogs and Social Media (ICWSM), 2010.
30.J. Leskovec, L. A. Adamic, and B. A. Huberman. The dynamics of viral marketing. In EC ’06: Proceedings of the 7th ACM conference on Electronic commerce, pages 228–237, New York, NY, USA, 2006. ACM.
31.D. Liben-Nowell and J. Kleinberg. Tracing information flow on a global scale using internet chain-letter data. Proceedings of the National Academy of Sciences, 105(12):4633, 2008.
32.C. F. Manski. Identification of endogenous social effects: The reflection problem. Rev. Econ. Stud., 60(3):531–42, July 1993.
33.A. Marin. Are respondents more likely to list alters with certain characteristics? Implications for name generator data. Social Networks, 26(4):289–307, Oct. 2004.
34.M. McPherson, L. S. Lovin, and J. M. Cook. Birds of a Feather: Homophily in Social Networks. Annu. Rev. Sociol., 27(1):415–444, 2001.
35.M. E. J. Newman. Spread of epidemic disease on networks. Phys. Rev. E, 66(1):016128, Jul 2002.
36.J.-P. Onnela and F. Reed-Tsochas. Spontaneous emergence of social influence in online systems. Proceedings of the National Academy of Sciences, 107(43):18375–18380, 2010.
37.J. P. Onnela, J. Saram¨aki, J. Hyv¨onen, G. Szab´o, D. Lazer, K. Kaski, J. Kert´esz, and A. L. Barab´asi. Structure and tie strengths in mobile communication networks. Proceedings of the National Academy of Sciences, 104(18):7332–7336, May 2007.
38.K. Purcell, L. Rainie, A. Mitchell, T. Rosenstiel, and K. Olmstead. Understanding the participatory news consumer. Technical report, Pew Internet & American Life Project, 2010.
39.C. R. Shalizi and A. C. Thomas. Homophily and Contagion Are Generically Confounded in Observational Social Network Studies. Sociological Methods and Research, 27:211–239, 2011.
40.T. Stein, E. Chen, and K. Mangla. Facebook Immune System. In EuroSys Social Network Systems, 2011.
41.E. S. Sun, I. Rosenn, C. A. Marlow, and T. M. Lento. Gesundheit! modeling contagion through facebook news feed. In Proceedings of the 3rd Int’l AAAI Conference on Weblogs and Social Media, San Jose, CA, 2009. AAAI.
42.D. J. Watts and S. H. Strogatz. Collective dynamics of ‘small-world’ networks. Nature, 393(6684):440–442, June 1998.
43.S. Wu, J. M. Hofman, W. A. Mason, and D. J. Watts. Who says what to whom on twitter. In ACM Conference on the World Wide Web, Hyderbad, India, 2011. ACM Press.

Citation

MLA

Bakshy, E., et al. “The Role of Social Networks in Information Diffusion”. Proceedings of the 21st International Conference on World Wide Web, 2012, pp. 519–28, https://doi.org/10.1145/2187836.2187907.

APA

Bakshy, E., Rosenn, I., Marlow, C., & Adamic, L. (2012). The role of social networks in information diffusion. Proceedings of the 21st International Conference on World Wide Web, 519–528. https://doi.org/10.1145/2187836.2187907

Chicago

Bakshy, E., I. Rosenn, C. Marlow, and L. Adamic. 2012. “The Role of Social Networks in Information Diffusion”. Proceedings of the 21st International Conference on World Wide Web, 519–28. https://doi.org/10.1145/2187836.2187907.

Harvard

Bakshy, E. et al. (2012) “The role of social networks in information diffusion”, Proceedings of the 21st international conference on World Wide Web. ACM, pp. 519–528. Available at: https://doi.org/10.1145/2187836.2187907.

Vancouver

1. Bakshy E, Rosenn I, Marlow C, Adamic L (2012) The role of social networks in information diffusion. In: Proceedings of the 21st international conference on World Wide Web. ACM, pp 519–528

BibTeX

@inproceedings{Bakshy_2012, series={WWW 2012}, title={The role of social networks in information diffusion}, url={http://dx.doi.org/10.1145/2187836.2187907}, DOI={10.1145/2187836.2187907}, booktitle={Proceedings of the 21st international conference on World Wide Web}, publisher={ACM}, author={Bakshy, Eytan and Rosenn, Itamar and Marlow, Cameron and Adamic, Lada}, year={2012}, month=Apr, pages={519–528}, collection={WWW 2012} }

Metadata:Crossref

Access the Paper

This paper is available from its original source. Click below to access the PDF.

Open PDF

The role of social networks in information diffusion

Abstract

Topics

Keywords

Table of Contents

Knowls

Knowl 1 — Randomized In Situ Feed Experiment for Identifying Social Influence

Knowl 2 — Overall Causal Impact of Feed Exposure on Information Sharing

Knowl 3 — Collective Dominance of Weak Ties in Aggregate Information Diffusion

Knowl 4 — Formulation of Collective Tie-Strength Influence

Knowl 5 — Individual Tie Strength, Influence, and Information Novelty

Knowl 6 — Differential Absolute and Relative Effects of Multiple Sharing Friends

Knowl 7 — Interaction-Based Measures of Tie Strength

Knowl 8 — Feed Acceleration of Information Diffusion and Baseline Temporal Clustering

Knowl 9 — Non-Feed Observational Bounds on Homophily and Influence

Knowl 10 — Experimental Population Demographics and Balance

References

Citation

BibTeX

Access the Paper