Abstract:
A criterion based on mutual information among variables is proposed for building a bayesian tree from a finite number of examples. The factor graph, in Forney-style form,...Show MoreMetadata
Abstract:
A criterion based on mutual information among variables is proposed for building a bayesian tree from a finite number of examples. The factor graph, in Forney-style form, can be used as an associative memory that performs probabilistic inference in data fusion applications. The procedure is explained with the aid of a fully-described example.
Date of Conference: 14-16 June 2010
Date Added to IEEE Xplore: 14 October 2010
ISBN Information: