ABSTRACT
A bi-type heterogeneous relational data set consists of two types of data objects with heterogeneous relations between them. Bi-type heterogenous relational data are a very important special case of heterogeneous relational data, since they arise frequently in various important applications. In bi-type heterogeneous relational data clustering, we are interested in clustering two types of data objects simultaneously. This is also known as co-clustering in the literature. In this chapter, we present a new co-clustering framework, Block Value Decomposition (BVD), for bi-type heterogeneous relational data, which factorizes the relational data matrix into three components: the row-coefficient matrix R, the block value matrix B, and the column-coefficient matrix C.
