Model Construction Algorithms for Object-Oriented Probabilistic Relational Models

This paper presents three new algorithms for the automatic construction of models from Object Oriented Probabilistic RelationalModels. The first two algorithms are based on the knowledge based model construction approach while the third is based on an Object Oriented Bayesian Network instance tree triangulation method. We discuss the strengths and limitations of each of the algorithms and compare their performance against the knowledge based model construction and Structured Variable Elimination algorithms developed for Probabilistic Relational Models.

• Type: Conference
• Year: 2006
• Where: FLAIRS
• Authors: Catherine Howard, Markus Stumptner
• Keywords: Algorithms, Artificial Intelligence, FLAIRS 2006, Knowledge Based Model, Object Oriented