MVMC is multiview data clustering algorithm based on multiobjective evolutionary optimization, where the multiview property refers to the availability of multiple feature sets and/or multiple relational descriptions. The approach takes advantage of many-objective optimization concepts to explore a range of (Pareto optimal) trade-offs, while scaling to settings with three or more data views.

MVMC is described in detail in our paper:

A. José-García, J. Handl, W. Gómez-Flores, and M. Garza-Fabre
An Evolutionary Many-objective Approach to Multiview Clustering Using Feature and Relational Data
Applied Soft Computing
https://doi.org/10.1016/j.asoc.2021.107425
[see the attached PDF file: ASOC_manuscript.pdf]

MVMC was developed with MATLAB R2020b. To try the algorithm look at the scripts demo_mvmc.m and mvmc_experiments.m.

Contact:

Adán José-García (adanjoga@gmail.com)
Julia Handl (julia.handl@manchester.ac.uk)
Wilfrido Gómez-Flores (wgomez@cinvestav.mx)
Mario Garza-Fabre (garzafabre@gmail.com)