Clustering Social Networks

Mishra, Nina; Schreiber, Robert; Stanton, Isabelle; Tarjan, Robert E.

doi:10.1007/978-3-540-77004-6_5

Cited by 98 publications

(77 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mishra et al [12] have used spectral clustering for social network analysis in 2007. They aimed at finding good cuts on the basis of conductance, i.e.…”

Section: Related Workmentioning

confidence: 99%

On Spectral Partitioning of Co-authorship Networks

Snåšel

Krömer

Platoš

et al. 2012

Computer Information Systems and Industrial Management

View full text Add to dashboard Cite

Abstract. Spectral partitioning is a well known method in the area of graph and matrix analysis. Several approaches based on spectral partitioning and spectral clustering were used to detect structures in real world networks and databases. In this paper, we explore two community detection approaches based on the spectral partitioning to analyze a coauthorship network. The partitioning exploits the concepts of algebraic connectivity and characteristic valuation to form components useful for the analysis of relations and communities in real world social networks.

show abstract

“…Mishra et al [12] have used spectral clustering for social network analysis in 2007. They aimed at finding good cuts on the basis of conductance, i.e.…”

Section: Related Workmentioning

confidence: 99%

On Spectral Partitioning of Co-authorship Networks

Snåšel

Krömer

Platoš

et al. 2012

Computer Information Systems and Industrial Management

View full text Add to dashboard Cite

show abstract

“…Clustering techniques have been used to identify communities and study their evolution over time (Mishra et al, 2007). An important property found in many networks is community structure, in which network nodes are joined together in tightly knit groups, between which there are only looser connections.…”

Section: Social Network Analysismentioning

confidence: 99%

Sensor-based organisational design and engineering

Olguín-Olguín

Pentland

2010

IJODE

View full text Add to dashboard Cite

Abstract:We propose a sensor-based organisational design and engineering approach that combines behavioural sensor data with other sources of information such as e-mail, surveys, and performance data in order to design interventions aimed at improving organizational outcomes. We discuss relevant theory and technology backgrounds and describe the general requirements of a sensor-based organizational design and engineering system. We present an experimental platform that combines sensor measurements, pattern recognition algorithms, statistical analysis, social network analysis, and feedback mechanisms to study the relationship between social signalling behaviour and face-to-face interaction networks, with job attitudes and performance. We describe three case studies that we have conducted in several organisations using our experimental platform and the methodology that we have followed. The first study looks at e-mail and face-to-face networks in a marketing division of a bank. The second study analyses the effects of nurses" social behaviour on patients" length of stay in the postanaesthesia care unit of a hospital. Finally, the third study analyses the effects of retail bank employees" social behaviour on sales performance. Keywords: Organisational design; organisational engineering; organisational behaviour; sociometric badges; social sensors; interventions.Reference to this paper should be made as follows: Olguín Olguín, D. and Pentland, A. D. Olguín Olguín and A. PentlandAlex (Sandy) Pentland is a pioneer in wearable computers, mobile information systems, technology for developing countries, consumer health, and computational social science. He is one of the most-cited computer scientists in the world, and has won numerous international awards in the Arts, Sciences and Engineering. He is a co-founder of the Wearable Computing research community, the Autonomous Mental Development research community, the Center for Future Health, and was the founding director of the Media Lab Asia. He was formerly the Academic Head of the MIT Media Laboratory and is now MIT"s Toshiba Professor of Media Arts and Sciences. He is the founder and current faculty director of the Human Dynamics group. Pentland"s focus is on the development of human-centred technology, and the creation of ventures that take this technology into the real world. His work provides people with a clearer picture of their social environments, and helps companies and communities to reinvent themselves to be both more human and more productive.

show abstract

“…In some settings, like discovering communities in social networks [29,22], the clusters are naturally overlapping and by restricting our attention to non-overlapping clustering, we may lose valuable information about the structure of communities in a social network [29]. For example, consider a graph with a small number of popular nodes that are well-connected to many other nodes in the graph.…”

Section: Introductionmentioning

confidence: 99%

“…These nodes may naturally belong to more than one cluster. For more applications that show that overlapping clustering is more suitable than non-overlapping clustering, see clustering for social networks [29,22], clustering for distributed computing [28,3], clustering for inherent multi-assignment clustering [36] and clustering large networks for distributed PageRank computation and performing distributed random walks [3]. For a survey on such models of graph clustering, see the article by Brandes et al [11] and the references therein.…”

Section: Introductionmentioning

confidence: 99%

“…For example, Arora et al [4] study the problem of overlapping communities, present an axiomatic approach for defining overlapping social communities, and design approximation algorithms for discovering such communities in a set of randomly generated graphs. Moreover, Balcan et al [8] considered the problem of enumerating and identifying communities generalizing and refining the definition of communi-ties by Mishra et al [29]. These papers do not explicitly attempt to optimize the cut conductance, but they consider other related notions for quality of clusters like the cut value and density of clustersl [4,8].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On the Advantage of Overlapping Clusters for Minimizing Conductance

2013

View full text Add to dashboard Cite

Graph clustering is an important problem with applications to bioinformatics, community discovery in social networks, distributed computing, and more. While most of the research in this area has focused on clustering using disjoint clusters, many real datasets have inherently overlapping clusters. We compare overlapping and non-overlapping clusterings in graphs in the context of minimizing their conductance. It is known that allowing clusters to overlap gives better results in practice. We prove that overlapping clustering may be significantly better than non-overlapping clustering with respect to conductance, even in a theoretical setting.For minimizing the maximum conductance over the clusters, we give examples demonstrating that allowing overlaps can yield significantly better clusterings, namely, one that has much smaller optimum. In addition for the min-max variant, the overlapping version admits a simple approximation algorithm, while our algorithm for the non-overlapping version is complex and yields a worse approximation ratio due to the presence of the additional constraint. Somewhat surprisingly, for the problem of minimizing the sum of conductances, we found out that allowing overlap does not help. We show how to apply a general technique to transform any overlapping clustering into a non-overlapping one with only a modest increase in the sum of conductances. This uncrossing technique is of independent interest and may find further applications in the future.We consider this work as a step toward rigorous comparison of overlapping and nonoverlapping clusterings and hope that it stimulates further research in this area. * IBM T.J.Watson research center.

show abstract

Clustering Social Networks

Cited by 98 publications

References 16 publications

On Spectral Partitioning of Co-authorship Networks

On Spectral Partitioning of Co-authorship Networks

Sensor-based organisational design and engineering

On the Advantage of Overlapping Clusters for Minimizing Conductance

Contact Info

Product

Resources

About