A survey of association rule classification algorithms in data mining

Abstract

Data mining plays a vital role in discovering hidden patterns and unknown knowledge from different types of data bases. Association rule mining is not finding specific classes instead it identifies the frequent items but in classification, classifiers are used to determine specific classes. Integrating these two techniques gives more efficient approach called Associative Classification. It is a new era in data mining approaches which is integrating Association rule and Classification to build accurate classifier than traditional methods. Most of the researchers proved that AC produces accurate results and also time efficient with different datasets. There are several algorithms proposed in recent times for associative classification (AC) such as Classification based on Association (CBA), Classification based on Multiple Association Rules (CMAR) and Classification based Predictive Association Rule (CPAR). This study compares and analyses the various important AC algorithms in terms of method, contributions, experimental results, accuracy and execution time irrespective of data sets.