I. Stajduhar (Croatia) and I. Bratko (Slovenia)
machine learning, probabilistic networks, classification
Learning directed probabilistic networks from data and us ing them for classification purposes is a well known prob lem. Many learning algorithms have been shown to be suc cessful for various kinds of learning scenarios. Basically they all generate a single network from data, which is then used for classification purposes and possible domain un derstanding. In this paper we propose a simple method for inferring a model consisting of several Bayesian networks, each one representing data of one class. The data is di vided into class subsets and from each subset a separate Bayesian network is learnt. Classification is done using prior and posterior probability distribution information in all networks. We thoroughly tested the proposed method on synthetic data and several repository datasets and com pared it to other machine learning methods, to prove its effectiveness. We argue that with smaller modifications, the method can be used for learning from censored survival domains.
Important Links:
Go Back
