Statistical Relational Learning (SRL), studies techniques that combine the
strengths of relational learning (e.g.
inductive logic programming) and
probabilistic learning
(e.g. Bayesian networks). By combining the power of logic and probability,
such systems can perform robust and accurate reasoning and learning about
complex relational data. See the book:
Introduction to Statistical Relational
Learning. Our work in the area has primarily focused on applications of
SRL methods to problems in
natural language processing,
transfer learning, and
abductive reasoning.