On Wednesday, January 31st, 2024, the seminar series hosted a talk by Dr. Joshua Tobin, a postdoctoral researcher in Trinity College Dublin, discussing co-clustering multi-view data with the latent block model. Details for the talk are below.
Title
Co-clustering multi-view data with the Latent Block Model
Abstract
The Latent Block Model (LBM) is a prominent model-based co-clustering method, returning parametric representations of each block cluster. The LBM, while adapted in literature to handle varying feature types, cannot represent information which is spread heterogeneously between data views. Motivated by the analysis of a multi-modal public health dataset, we introduce the multi-view LBM, extending the LBM method to multi-view data where each view marginally follows a LBM. In the case of two views, the dependence between them is captured by a cluster membership matrix and we aim to learn the structure of this matrix. We develop a likelihood-based approach for parameter estimation and describe penalized methods to encourage sparsity in the cluster membership matrix. To verify the connection between the multiple data views, we extend recent work developing tests for the null hypothesis that the latent row-cluster memberships between the views are independent. This testing procedure is integrated into the estimation strategy, yielding a method that is useful for exploring multiple high-dimensional data sets.