ACNB: Associative Classification Mining Based on Naieve Bayesian Method

摘要

Integrating association rule discovery and classification in data mining brings a new approach known as associative classification. Associative classification is a promising approach that often constructs more accurate classification models (classifiers) than the traditional classification approaches such as decision trees and rule induction. In this research, the authors investigate the use of associative classification on the high dimensional data in text categorization. This research focuses on prediction, a very important step in classification, and introduces a new prediction method called Associative Classification Mining based on Naive Bayesian method. The running time is decreased by removing the ranking procedure that is usually the first step in ranking the derived Classification Association Rules. The prediction method is enhanced using the Naive Bayesian Algorithm. The results of the experiments demonstrate high classification accuracy.