Multi-View Data Learning with Deep Matrix Factorization

PhD Project details

Title: Multi-View Data Learning with Deep Matrix Factorization

Due to the rapid development in technology as well as the extensive amount of storage space available, many data are often collected from multiple sources and therefore cannot be expressed using a single view. Thus, many real-world datasets illustrate data using multiple representations or views, where each view captures different heterogeneous angles for the given information. Learning meaningful information directly from multi-view data is problematic due to characteristics such as high dimensionality, sparsity, and the possibility of containing redundant information. To deal with these problems, most of the multi-view data learning methods first reduce the dimensionality from the original high-dimensional space to low-dimensional space and then use the learned latent features to perform subsequent machine learning tasks such as clustering and classification. Among various dimensionality reduction techniques, deep matrix factorization-based methods have recently been explored in multi-view data learning due to their ability to deal with complex non-linear data. This research endeavors to design various deep matrix factorization-based models for effective feature learning.

Supervisory team:

Prof. Richi Nayak,  Dr Thirunavukarasu Balasubramaniam, Dr Khanh Luong