Biometrical Letters vol. 46(2), 2009, pp. 129-152
|
LEARNING BAYESIAN NETWORKS USING EXPERT'S PRIOR INFORMATION ON STRUCTURES Massimiliano Mascherini1, Alessandro Camussi2, Federico M. Stefanini3 1Joint Research Centre of the European Commission, 21027 Ispra (Va), Italy, e-mail: massimiliano.mascherini@jrc.it 2Department of Agricultural Biotechnology, Genetics Unit, University of Florence, via Maragliano, 75-77 50144 Florence, Italy, e-mail: alessandro.camussi@unifi.it 3Department of Statistics, University of Florence, viale Morgagni, 59 50134 Florence, Italy, e-mail: stefanini@ds.unifi.it |
Most of the approaches developed in the literature to elicit the a priori distribution on Directed Acyclic Graphs (DAGs) require a full specification of graphs. Nevertheless, expert's prior knowledge about conditional independence relations may be weak, making the elicitation task troublesome. This paper presents and evaluates an elicitation procedure for DAGs which exploits prior knowledge on network topology. The elicitation is suited to large Bayesian Networks (BNs) and it accounts for immediate causal link and DAG sparsity. We develop a new quasi-Bayesian score function, the P-metric, to perform structural learning following a score-and-search approach. We tested our score function on two different benchmark BNs by varying sample size and prior belief concerning structures. Our results show the effectiveness of the proposed method and suggest that the use of prior information improves the structural learning process.
Bayesian Networks; Structural Learning; Prior Information
