A Study of Breast Cancer Detection for Various Classification Techniques

Breast cancer is one of the common diseases for women. Cancer is an abnormal growth of cells that can be either precancer or dangerous stages. This research to detect the breast cancer using classification techniques The first stage is preprocessing on the data is done from the Wisconsin dataset from UCI machine learning .In this experiment compare three classification techniques and comparison results show that Naïve bayes has higher prediction accuracy i.e. 97.4% than SMO and K star methods. The next stage is reducing the dimension of breast cancer database from ten to nine by using and sorting the attributes by using feature selection method is used to improve the accuracy. After applied feature selection method, the results show that that Naive bayes has higher prediction accuracy i.e. 97.80% than SMO and K star methods. All this experiment is done by WEKA software.