Collaborative Generative Topographic Mapping

Mohamad Ghassany, Nistor Grozavu, Younès Bennani

Abstract: The aim of collaborative clustering is to reveal the common structure of data distributed on different sites. In this paper, we present a new approach for the topological collaborative clustering using a generative model, which is the Generative Topographic Mappings (GTM). In this case, maps representing different sites could collaborate without recourse to the original data, preserving their privacy. Depending ont the data structure, there are three different ways of collaborative clustering: horizontal, vertical and hybrid. In this study we introduce the Collaborative GTM for the vertical collaboration. The article presents the formalism of the approach and its validation. The proposed approach has been validated on several datasets and experimental results have shown very promising performance.