Performance evaluation of compromise conditional Gaussian networks for data clustering

This paper is devoted to the proposal of two classes of compromise conditio nal Gaussian networks for data clustering as well as to their experimental evaluation and comparison on synthetic and real-world databases. According to the reported results, the models show an ideal trade-off between efficie ncy and effectiveness, i.e., a balance between the cost of the unsupervised model learning process and the quality of the learnt models. Moreover, the proposed models are very appealing due to their closeness to human intuiti on and computational advantages for the unsupervised model induction proces s. while preserving a rich enough modeling power. (C) 2001 Elsevier Science Inc. All rights reserved.