A fuzzy associative rule based approach for pattern mining and pattern based classification

Abstract

Associative Classification leverages Association Rule Mining (ARM) to train Rule-based classifiers. The classifiers are built on high quality Association Rules mined from the given dataset. Associative Classifiers are very accurate because Association Rules encapsulate all the dominant and statistically significant relationships between items in the dataset. They are also very robust as noise in the form of insignificant and low-frequency itemsets are eliminated during the mining and training stages. Moreover, the rules are easy-to-comprehend, thus making the classifier transparent. newline newlineIn this thesis we have described a broad framework for Associative Classification, especially Fuzzy Associative Classification, which starts from the pre-processing (in case of fuzzy classification) of the original dataset to an appropriate fuzzy format for ARM. Then, association rules are mined using our ARM algorithms. These association rules are then leveraged to build an Associative Classifier (fuzzy or crisp). This is done using our associative classification algorithms. Last, this approach can be applied to build associative classifiers in specialized domains. This can be done using customized associative classification algorithms, which are based on our associative classification approach and framework.