Community Detection Using Diffusion Information

Ramezani, Maryam; Khodadadi, Ali; Rabiee, Hamid R.

doi:10.1145/3110215

Cited by 28 publications

(29 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Social behaviour dynamics, using information diffusion and probabilistic framework, have been further explored using maximum likelihood estimation [105]. Infor-mation diffusion have been also applied to detect communities without prior knowledge of the underlying network [81]. Expectation maximization based approaches have been explored to infer the spread of information on social networks [10,11].…”

Section: Information Diffusion-based Communitiesmentioning

confidence: 99%

Community elicitation from co-occurrence of activities

Mengoni

Milani

Poggioni

et al. 2020

Future Generation Computer Systems

View full text Add to dashboard Cite

A massive amount of data is continuously generated by the activities that people hold on online platforms, mobile systems and in general using and interacting with digital systems. Big data, not directly related to social networks, are generated in large quantities from systems that are not necessarily social systems. In fact, in the information society a whole lot of activities take place on systems that were not developed to support explicit social interactions between users.In this thesis, starting from the observation of users activities within a system, we want to investigate if it is possible to characterise the existence of social relationships among them. As people interact, individually or in groups, we want to elicit their social communities from the temporal and spatial co-occurrence of their activities.The key assumption of this work is that we suppose that there are multiple, parallel, hidden communication channels and social networks where social interactions take place among users and which determine the observed emergent phenomenon of actions co-occurrences.The main original contribution of this thesis is the proposal of innovative methodologies for users community discovery from implicit user-system interactions and their experimental evaluation. The History Based Analysis approach is a novel approach we have introduced, that exploits the similarity of users' activity histories to discover the hidden social communities. To better characterise the histories binary correlation measures we have introduced and experimented original entropy i amplification factors that take in account system wide distribution of activities at a given time to contextualise the user activity similarities. The other relevant introduced approach, the Session Based method, uses graph based representation of concurrent users' sessions to elicit the hidden social communities. Both proposed approaches have been validated using a real world dataset containing the activity logs of students using a virtual learning environment platform.A remarkable result of our work has been to confirm that co-occurrence of people activities is an emerging epiphenomenon of hidden, implicit information exchanges through side channel communications. Therefore the observation of co-occurrence of events can be used to elicit social relationships. Interesting extension of this work include the analysis of real world co-occurrences, like in the case of people, personal vehicles or other personal objects occurring in the same physical place at the same time, and in general wherever it co-occurrence can be seen as an emerging epiphenomenon of people's relationships and information exchange.Potential applications of this thesis work can fall in various areas such as business, marketing, public administration, including intelligence and military sectors.Experimental evaluation of the introduced methodologies through tests held in the domain of eLearning demonstrated the effectiveness of our proposed approaches in retrieving hidden social communities.

show abstract

Section: Information Diffusion-based Communitiesmentioning

confidence: 99%

Community elicitation from co-occurrence of activities

Mengoni

Milani

Poggioni

et al. 2020

Future Generation Computer Systems

View full text Add to dashboard Cite

show abstract

“…The authors propose two algorithms: C-IC takes into account only participation of a node in a cascade; C-Rate includes the time stamps, but limits the node's influence by its own community. Recently, [26] proposed an alternative maximum likelihood approach, which exploits the Markov property of the cascades. As an input, similarity scores of node pairs are computed, based on their joint participation in cascades.…”

Section: Community Inference From Cascadesmentioning

confidence: 99%

“…As an input, similarity scores of node pairs are computed, based on their joint participation in cascades. The R-CoDi algorithm in [26] starts with a random partition, while D-CoDi starts with a partition obtained by DANI [27]. We use all four mentioned algorithms as our baselines.…”

Section: Community Inference From Cascadesmentioning

confidence: 99%

“…Finally, we used algorithms proposed for solving the same problem as in our work: C-IC and C-Rate [5], as well as R-CoDi and D-CoDi [26]. We use the publicly available implementations provided by the authors.…”

Section: Baselinesmentioning

confidence: 99%

“…In this paper, we analyze the problem, which recently attracted interest in the literature [5,26], of inferring community structure of a given network based solely on cascades (e.g., information or epidemic) propagating through this network. Communities are groups of highly interconnected nodes with relatively few edges joining nodes of different groups [9], such as groups by interests in social networks or pages on related topics on the Web.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning Clusters through Information Diffusion

Prokhorenkova

Tikhonov²,

Litvak

2019

The World Wide Web Conference

View full text Add to dashboard Cite

When information or infectious diseases spread over a network, in many practical cases, one can observe when nodes adopt information or become infected, but the underlying network is hidden. In this paper, we analyze the problem of finding communities of highly interconnected nodes, given only the infection times of nodes. We propose, analyze, and empirically compare several algorithms for this task. The most stable performance, that improves the current state-of-the-art, is obtained by our proposed heuristic approaches, that are agnostic to a particular graph structure and epidemic model.

show abstract

Effects of Hidden Users on Cascade-Based Community Detection

Suzuki

Tsugawa

2021

Complex Networks XII

View full text Add to dashboard Cite

Community Detection Using Diffusion Information

Cited by 28 publications

References 34 publications

Community elicitation from co-occurrence of activities

Community elicitation from co-occurrence of activities

Learning Clusters through Information Diffusion

Effects of Hidden Users on Cascade-Based Community Detection

Contact Info

Product

Resources

About