Context-based statistical relational learning

Abstract

The relational structure is an important source of information, which is often ignored by the traditional statistical learning methods. Thus this thesis focuses on how to explicitly exploit such relational information in statistical learning tasks so as to build more effective and more robust models. The main methodology used in the thesis is derived from context-based modeling and analysis. Several models and algorithms are investigated from different viewpoints of context, thereby demonstrating the general applicability of context-based statistical relational learning.