Abstract
The weight-based fusion model (WBFM) is one of the simplest and most efficient models for community detection (CD) in node-attributed social networks (ASNs) which contain both links between social actors (aka structure) and actors’ features (aka attributes). Although WBFM is widely used, it has a logical gap as we show here. Namely, the gap stems from the discrepancy between the so-called Composite Modularity that is usually optimized within WBFM and the measures used for CD quality evaluation. The discrepancy may cause the misinterpretation of CD results and difficulties with the parameter tuning within WBFM. To fulfil the gap, we theoretically study how Composite Modularity is related to the CD quality measures. This study further yields a pioneering non-manual parameter tuning scheme that provides the equal impact of structure and attributes on the CD results. Experiments with synthetic and real-world ASNs show that our conclusions help to reasonably interpret the CD results and that our tuning scheme is very accurate.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
An edge weight may be zero and this indicates that there is no social connection.
- 2.
If one deals with nominal or textual attributes, it is common to use one-hot encoding or embeddings to obtain their numerical representation.
- 3.
Communities may be overlapping if necessary but here we focus on disjoint ones.
References
Akbas, E., Zhao, P.: Graph clustering based on attribute-aware graph embedding. In: Karampelas, P., Kawash, J., Özyer, T. (eds.) From Security to Community Detection in Social Networking Platforms, pp. 109–131. Springer, Cham (2019)
Alinezhad, E., Teimourpour, B., Sepehri, M.M., Kargari, M.: Community detection in attributed networks considering both structural and attribute similarities: two mathematical programming approaches. Neural Comput. Appl. 32, 3203–3220 (2020)
Blondel, V.D., Guillaume, J.L., Lambiotte, R., Lefebvre, E.: Fast unfolding of communities in large networks. J. Stat. Mech. Theor. Exp. 2008(10), P10008 (2008)
Bothorel, C., Cruz, J., Magnani, M., Micenková, B.: Clustering attributed graphs: models, measures and methods. Netw. Sci. 3(3), 408–444 (2015)
Chunaev, P., Nuzhdenko, I., Bochenina, K.: Community detection in attributed social networks: a unified weight-based model and its regimes. In: 2019 International Conference on Data Mining Workshops (ICDMW), pp. 455–464 (2019)
Chunaev, P.: Community detection in node-attributed social networks: a survey. Comput. Sci. Rev. 37, 100286 (2020)
Combe, D., Largeron, C., Egyed-Zsigmond, E., Gery, M.: Combining relations and text in scientific network clustering. In: Proceedings of the 2012 International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012, pp. 1248–1253 (2012)
Cruz, J., Bothorel, C., Poulet, F.: Entropy based community detection in augmented social networks. In: International Conference on Computational Aspects of Social Networks, pp. 163–168 (2011)
Cruz, J., Bothorel, C., Poulet, F.: Détection et visualisation des communautés dans les réseaux sociaux. Revue d’intelligence artificielle 26, 369–392 (2012)
Dang, T.A., Viennet, E.: Community detection based on structural and attribute similarities. In: International Conference on Digital Society, pp. 7–14 (2012)
McPherson, M., Smith-Lovin, L., Cook, J.M.: Birds of a feather: homophily in social networks. Ann. Rev. Sociol. 27(1), 415–444 (2001)
Meng, F., Rui, X., Wang, Z., Xing, Y., Cao, L.: Coupled node similarity learning for community detection in attributed networks. Entropy 20(6), 471 (2018)
Neville, J., Adler, M., Jensen, D.: Clustering relational data using attribute and link information. In: Proceedings of the Text Mining and Link Analysis Workshop, 18th International Joint Conference on Artificial Intelligence, pp. 9–15 (2003)
Newman, M.E.J., Girvan, M.: Finding and evaluating community structure in networks. Phys. Rev. E 69, 026113 (2004)
Ruan, Y., Fuhry, D., Parthasarathy, S.: Efficient community detection in large networks using content and links. In: Proceedings of the 22nd International Conference on World Wide Web, WWW 2013, pp. 1089–1098 (2013)
Steinhaeuser, K., Chawla, N.V.: Identifying and evaluating community structure in complex networks. Pattern Recogn. Lett. 31(5), 413–421 (2010)
Acknowledgements
This research was financially supported by the Russian Science Foundation, Agreement 19-71-10078.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Chunaev, P., Gradov, T., Bochenina, K. (2021). Composite Modularity and Parameter Tuning in the Weight-Based Fusion Model for Community Detection in Node-Attributed Social Networks. In: Benito, R.M., Cherifi, C., Cherifi, H., Moro, E., Rocha, L.M., Sales-Pardo, M. (eds) Complex Networks & Their Applications IX. COMPLEX NETWORKS 2020 2020. Studies in Computational Intelligence, vol 943. Springer, Cham. https://doi.org/10.1007/978-3-030-65347-7_9
Download citation
DOI: https://doi.org/10.1007/978-3-030-65347-7_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-65346-0
Online ISBN: 978-3-030-65347-7
eBook Packages: EngineeringEngineering (R0)