Learning Dynamic Bayesian Networks with the TOM4L Process

This paper addresses the problem of learning a Dynamic Bayesian Network from timed data without prior knowledge to the system. One of the main problems of learning a Dynamic Bayesian Network is building and orienting the edges of the network avoiding loops. The problem is more difficult when data are timed. This paper proposes a new algorithm to learn the structure of a Dynamic Bayesian Network and to orient the edges from the timed data contained in a given timed data base. This algorithm is based on an adequate representation of a set of sequences of timed data and uses an information based measure of the relations between two edges. This algorithm is a part of the Timed Observation Mining for Learning (TOM4L) process that is based on the Theory of the Timed Observations. The paper illustrates the algorithm with a theoretical example before presenting the results on an application on the Apache system of the ArcelorMittal Steel Group, a real world knowledge based system that diagnoses a galvanization bath.