Clustering methods for multi-aspect data

With the rapid growth of computational technology, multi-aspect data has become ubiquitous and popular. Multi-aspect data is able to represent the information with multiple perspectives or multiple types of features. This thesis explores this hot topic of machine learning and presents several methods of clustering by exploiting the multi-aspect data properties under the Non-negative Matrix Factorization framework with manifold learning. The proposed methods have shown superiority to identify subgroups for the high-dimensional, sparse and complex data. The proposed methods have applicability to wider fields such as vision, signal processing, bio-informatics, text mining, web mining and recommender systems