Bursty subgraphs in social networks

Eftekhar, Milad; Koudas, Nick; Ganjali, Yashar

doi:10.1145/2433396.2433423

Cited by 7 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The occurrence of a burst is modeled by an underlying state transiting into a bursty state that emits messages at a higher rate than at the non-bursty state. Based on this model, many variant models are proposed for detecting bursts from document streams [11], [38], e-commerce queries [12], time series [39], and social networks [13]. Although these models are theoretically interesting, some assumptions made by them are inappropriate, such as the Poisson process of message arrivals (see [40]) and nonexistence of spams/bots, which may limit their practical usage.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Tracking triadic cardinality distributions for burst detection in high-speed graph streams

et al. 2021

View full text Add to dashboard Cite

In everyday life, we often observe unusually frequent interactions among people before or during important events, e.g., people send/receive more greetings to/from their friends on holidays than regular days. We also observe that some videos or hashtags suddenly go viral through people's sharing on online social networks (OSNs). Do these seemingly different phenomena share a common structure? All these phenomena are associated with sudden surges of user interactions in networks, which we call "bursts" in this work. We uncover that the emergence of a burst is accompanied with the formation of triangles in some properly defined networks. This finding motivates us to propose a new and robust method to detect bursts on OSNs. We first introduce a new measure, "triadic cardinality distribution", corresponding to the fractions of nodes with different numbers of triangles, i.e., triadic cardinalities, in a network. We show that this distribution not only changes when a burst occurs, but it also has a robustness property that it is immunized against common spamming socialbot attacks. Hence, by tracking triadic cardinality distributions, we can more reliably detect bursts than simply counting interactions on OSNs. To avoid handling massive activity data generated by OSN users during the triadic tracking, we design an efficient "sample-estimate" framework to provide maximum likelihood estimate on the triadic cardinality distribution. We propose several sampling methods, and provide insights about their performance difference, through both theoretical analysis and empirical experiments on real world networks.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Similar problem also exists when detecting bursts caused by usercontent interactions. Many previous works on burst detection are based on idealized assumptions [10], [11], [12], [13] and simply ignore the existence of social bots.…”

Section: Introductionmentioning

confidence: 99%

Tracking triadic cardinality distributions for burst detection in high-speed graph streams

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The occurrence of a burst is modeled by an underlying state transiting into a bursty state that emits messages at a higher rate than at the non-bursty state. Based on this model, many variant models are proposed for detecting bursts from document streams [39,22], e-commerce queries [24], time series [41], and social networks [13]. Although these models are theoretically interesting, some assumptions made by them are inappropriate, such as the Poisson process of message arrivals (see [4]) and nonexistence of spams/bots, which may limit their practical usage.…”

Section: Related Workmentioning

confidence: 99%

“…Similar problem also exist when detecting bursts caused by user-content interactions. Many previous works on burst detection are based on idealistic assumptions [17,39,24,13] and simply ignore the existence of social bots. Present work.…”

Section: Introductionmentioning

confidence: 99%

Tracking Triadic Cardinality Distributions for Burst Detection in Social Activity Streams

Zhao

Lui

Towsley³

et al. 2015

Proceedings of the 2015 ACM on Conference on Online Social Networks

View full text Add to dashboard Cite

In everyday life, we often observe unusually frequent interactions among people before or during important events, e.g., we receive/send more greetings from/to our friends on Christmas Day, than usual. We also observe that some videos suddenly go viral through people's sharing in online social networks (OSNs). Do these seemingly different phenomena share a common structure?All these phenomena are associated with sudden surges of user activities in networks, which we call "bursts" in this work. We find that the emergence of a burst is accompanied with the formation of triangles in networks. This finding motivates us to propose a new method to detect bursts in OSNs. We first introduce a new measure, "triadic cardinality distribution", corresponding to the fractions of nodes with different numbers of triangles, i.e., triadic cardinalities, within a network. We demonstrate that this distribution changes when a burst occurs, and is naturally immunized against spamming social-bot attacks. Hence, by tracking triadic cardinality distributions, we can reliably detect bursts in OSNs. To avoid handling massive activity data generated by OSN users, we design an efficient sample-estimate solution to estimate the triadic cardinality distribution from sampled data. Extensive experiments conducted on real data demonstrate the usefulness of this triadic cardinality distribution and the effectiveness of our sample-estimate solution.

show abstract

“…Micro-blogging platforms have also been used to monitor trends with novel applications such as predicting stock prices [23]. They have also been used to detect communities based on interests [14] or bursts [11] and to rank users based on their influence [27] within their community or based on their topical expertise [21]. Behavior of users on the social platforms and communities has also been studied [1,19].…”

Section: Related Workmentioning

confidence: 99%

Price trade-offs in social media advertising

Eftekhar

Thirumuruganathan

Das

et al. 2014

Proceedings of the Second ACM Conference on Online Social Networks

Self Cite

View full text Add to dashboard Cite

The prevalence of social media has sparked novel advertising models, vastly different from the traditional keyword based bidding model adopted by search engines. One such model is topic based advertising, popular with micro-blogging sites. Instead of bidding on keywords, the approach is based on bidding on topics, with the winning bid allowed to disseminate messages to users interested in the specific topic.Naturally topics have varying costs depending on multiple factors (e.g., how popular or prevalent they are). Similarly users in a micro-blogging site have diverse interests. Assuming one wishes to disseminate a message to a set V of users interested in a specific topic, a question arises whether it is possible to disseminate the same message by bidding on a set of topics that collectively reach the same users in V albeit at a cheaper cost.In this paper, we show how an alternative set of topics R with a lower cost can be identified to target (most) users in V. Two approximation algorithms are presented to address the problem with strong bounds. Theoretical analysis and extensive quantitative and qualitative experiments over real-world data sets at realistic scale containing millions of users and topics demonstrate the effectiveness of our approach.

show abstract

Bursty subgraphs in social networks

Cited by 7 publications

References 15 publications

Tracking triadic cardinality distributions for burst detection in high-speed graph streams

Tracking triadic cardinality distributions for burst detection in high-speed graph streams

Tracking Triadic Cardinality Distributions for Burst Detection in Social Activity Streams

Price trade-offs in social media advertising

Contact Info

Product

Resources

About