Learning Directed Probabilistic Logical Models using Ordering-search

Daan Fierens, Katholieke Universiteit Leuven, Belgium

Jan Ramon, Katholieke Universiteit Leuven, Belgium

Maurice Bruynooghe, Katholieke Universiteit Leuven, Belgium

Hendrik Blockeel, Katholieke Universiteit Leuven, Belgium

There is an increasing interest in upgrading Bayesian networks to the relational case resulting in so-called directed probabilistic logical models. In this paper we discuss how to learn non-recursive directed probabilistic logical models from relational data. This problem has already been tackled before by upgrading the structure-search algorithm for learning Bayesian networks. In this paper we propose to upgrade another algorithm, namely ordering-search, since for Bayesian networks this was found to work better than structure-search. In future work we plan to experimentally compare both approaches.