AClass: Classification algorithm based on association rule mining

Recent developments in data collecting technologies have led very large scale data warehouses that can be used effectively by decision makers. As long as basket type data such as customer buying attitudes are concerned, association rule mining is a useful and widely used method to extract patterns from large sets of data. Its tendency to find all possible relations between data fields makes it functional in many research domains; however it happens to be useless in domains like medicine and health as it produce lots of ineffectual rules concerning irrelevant data fields. Thus classifying a disease by using classic rule mining algorithms is not an easy task. Typical classification algorithms, on the other hand, only generate decision trees or classifiers according to pre-determined target; therefore, they need to be tuned to produce human readable rules that can be used in decision support. Another fact about data mining studies is their time wasting attitudes while working with large data sets. Thus just recently, parallel algorithms have attracted various authors. AClass algorithm is developed to integrate these two approaches so that simple classification rules can be generated with the power of association rule mining. Apriori algorithm is used as a base model and modified to be able to generate human readable classification association rules. As a final task, parallelization techniques were also used to show the ability of AClass algorithm to handle huge data sets. Results from both accuracy and parallelization tests are presented.