2009 | OriginalPaper | Buchkapitel
Convex Mixture Models for Multi-view Clustering
verfasst von : Grigorios Tzortzis, Aristidis Likas
Erschienen in: Artificial Neural Networks – ICANN 2009
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Data with multiple representations (views) arise naturally in many applications and multi-view algorithms can substantially improve the classification and clustering results. In this work, we study the problem of multi-view clustering and propose a multi-view convex mixture model that locates exemplars (cluster representatives) in the dataset by simultaneously considering all views. Convex mixture models are simplified mixture models that exhibit several attractive characteristics. The proposed algorithm extends the single view convex mixture models so as to handle data with any number of representations, taking into account the diversity of the views while preserving their good properties. Empirical evaluations on synthetic and real data demonstrate the effectiveness and potential of our method.