As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
Restricted Bayesian classifiers have demonstrated remarkable classification performance for data mining. However, the restricted network structure makes it impossible to represent the Markov blanket of class variable, which corresponds to the optimal classifier. And the test instances are not fully utilized, the final decision thus may be biased. In this paper, a novel unrestricted k-dependence classifier is proposed based on identifying the Markov blanket of the class variable. Furthermore, the algorithm also adopts local learning to build local structure, which can represent the evidence introduced by test instance. 15 datasets are selected from the UCI machine learning repository for zero-one loss comparison. The experimental results indicate that the unrestricted Bayesian classifier can achieve good trade-off between structure complexity and prediction performance.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.