Latent tree classifier
Y. Wang, N.L. Zhang, T. Chen, Leonard K. M. Poon · 2011
Abstract. We propose a novel generative model for classification called latenttreeclassifier(LTC).AnLTCrepresentseachclass-conditionaldistribution of attributes using a latent tree model, and uses Bayes rule to make prediction. Latent tree models can capture complex relationship among attributes. Therefore, LTC can approximate the true distribution behind data well and thus achieve good classification accuracy. We present an algorithm for learning LTC and empirically evaluate it on 37 UCI data sets. The results show that LTC compares favorably to the state-of-the-art. We also demonstrate that LTC can reveal underlying concepts and discover interesting subgroups within each class.