ABSTRACT
In more general cases, heterogeneous relational data consist of more than two types of data objects. Those multiple-type interrelated data objects formulates a k-partite heterogeneous relation graph. The research on mining the hidden structures from a k-partite heterogeneous relation graph is still limited and preliminary. In this chapter, we propose a general model, the relation summary network, to find the hidden structures (the local cluster structures and the global community structures) from a k-partite heterogeneous relation graph. The model provides a principal framework for unsupervised learning on kpartite heterogeneous relation graphs of various structures.
