2009 | OriginalPaper | Buchkapitel
A Statistical Approach for Binary Vectors Modeling and Clustering
Erschienen in: Advances in Knowledge Discovery and Data Mining
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
This paper presents an approach for Binary feature selection. Our selection technique is based on a principled statistical model using a finite mixture of distributions. In contrast with classic feature selection algorithms that have been proposed in supervised settings, where training data are available and completely labeled, our approach is fully unsupervised. Through some applications, we found that our feature selection model improves the clustering results.