Comparison of Classification Algorithms for Predicting Breast Cancer

Data mining is the process of extracting knowledge hidden in large volumes of data. Data mining tools predict future trends and behavior for knowledge driven decisions. This paper deals with predicting Breast Cancer using classification algorithms. Breast Cancer has been determined to be the second leading cause of death due to cancer and the most common serious type of cancer in women. The classification algorithms that we have taken for our research are J48, Naïve Bayes, LMT, REP Tree, Decision Table, K star, Simple logistic, Iterative classifier optimizer, IBK and Filtered classifier. Experiments were conducted on Wisconsin Breast Cancer datasets using machine learning tool WEKA with classification algorithms listed above. The metrics for the evaluation of the performance of various classification algorithm are accuracy and time taken for classification. The observation of the results have lead to the conclusion that J48 and Filtered classifier are good when it pertains to classification accuracy, although Filtered classifier consumes less time in comparison to J48. This has in fact opened up a plethora of interesting conclusions.