Self Weighted Multiplex Modularity Maximization for Multiview Clustering

Noureddine Henka, Mohamad Assaad, Sami Tazi
Proceedings of the 15th Asian Conference on Machine Learning, PMLR 222:406-421, 2024.

Abstract

In response to the challenge of representing data from multiple sources, researchers have proposed the use of multiplex graphs as a solution. Multiplex graphs are particularly useful for representing multi-view data, where each layer represents a specific type of interaction. Pillar community detection of multiplex graphs is a clustering application that computes groups of vertices across all layers. Modularity maximization is a popular technique for graph clustering, which has been generalized to multiplex graphs. However, this generalization did not consider the importance of each layer in pillar clustering. This paper presents a new technique called Self Weighted Multiplex Modularity (SWMM), which optimizes the weights associated with each layer and the partition that maximizes the multiplex modularity. The paper proposes two optimization methods, iterative and direct, and demonstrates the effectiveness and robustness of the technique in accurately retrieving clusters even when data is highly missing.

Cite this Paper

BibTeX
@InProceedings{pmlr-v222-henka24a, title = {Self Weighted Multiplex Modularity Maximization for Multiview Clustering}, author = {Henka, Noureddine and Assaad, Mohamad and Tazi, Sami}, booktitle = {Proceedings of the 15th Asian Conference on Machine Learning}, pages = {406--421}, year = {2024}, editor = {Yanıkoğlu, Berrin and Buntine, Wray}, volume = {222}, series = {Proceedings of Machine Learning Research}, month = {11--14 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v222/henka24a/henka24a.pdf}, url = {https://proceedings.mlr.press/v222/henka24a.html}, abstract = {In response to the challenge of representing data from multiple sources, researchers have proposed the use of multiplex graphs as a solution. Multiplex graphs are particularly useful for representing multi-view data, where each layer represents a specific type of interaction. Pillar community detection of multiplex graphs is a clustering application that computes groups of vertices across all layers. Modularity maximization is a popular technique for graph clustering, which has been generalized to multiplex graphs. However, this generalization did not consider the importance of each layer in pillar clustering. This paper presents a new technique called Self Weighted Multiplex Modularity (SWMM), which optimizes the weights associated with each layer and the partition that maximizes the multiplex modularity. The paper proposes two optimization methods, iterative and direct, and demonstrates the effectiveness and robustness of the technique in accurately retrieving clusters even when data is highly missing.} }
Endnote
%0 Conference Paper %T Self Weighted Multiplex Modularity Maximization for Multiview Clustering %A Noureddine Henka %A Mohamad Assaad %A Sami Tazi %B Proceedings of the 15th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Berrin Yanıkoğlu %E Wray Buntine %F pmlr-v222-henka24a %I PMLR %P 406--421 %U https://proceedings.mlr.press/v222/henka24a.html %V 222 %X In response to the challenge of representing data from multiple sources, researchers have proposed the use of multiplex graphs as a solution. Multiplex graphs are particularly useful for representing multi-view data, where each layer represents a specific type of interaction. Pillar community detection of multiplex graphs is a clustering application that computes groups of vertices across all layers. Modularity maximization is a popular technique for graph clustering, which has been generalized to multiplex graphs. However, this generalization did not consider the importance of each layer in pillar clustering. This paper presents a new technique called Self Weighted Multiplex Modularity (SWMM), which optimizes the weights associated with each layer and the partition that maximizes the multiplex modularity. The paper proposes two optimization methods, iterative and direct, and demonstrates the effectiveness and robustness of the technique in accurately retrieving clusters even when data is highly missing.
APA
Henka, N., Assaad, M. & Tazi, S.. (2024). Self Weighted Multiplex Modularity Maximization for Multiview Clustering. Proceedings of the 15th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 222:406-421 Available from https://proceedings.mlr.press/v222/henka24a.html.

Related Material