A comparison of classification efficiency of breast cancer patient datasets using data mining techniques

Abstract

The objectives of this research are: 1) to study a method for classifying breast cancer patient data, and 2) to compare the efficiency in breast cancer patient classification. Three data classification techniques, namely the k-nearest neighbor, decision tree, and naive bayes are used to test with the breast cancer patient dataset. The experimental results show that the classification of the naive bayes technique gives the best performance with 60% accuracy, 70% precision, 63% recall, and 57% f1-score. Followed by a decision tree that yields 60% accuracy, 61% precision, 61% recall, and 60% f1-score. Finally, the k-nearest neighbor method yields 46% accuracy, 23% precision, 50% recall, and 31% f1-score, respectively.