Improved Framework for Breast Cancer Prediction Using Frequent Itemsets Mining for Attributes Filtering

摘要

Data Mining is applicable for pulling up some new information by analyzing the database. It is also used for prediction based on real and actual current data. Breast cancer is a very harmful disease which effects badly to ones social, physical life and also effects mentally. This paper focuses on the attribute filtering techniques i.e frequent itemsets mining with the intention to find the essential and relevant attribute from the Wisconsin breast cancer dataset and classification algorithmic program like SVM, Naive Bayes, k-NN, Decision Tree comparison is done with attribute filtering. SVM produces beat result among all the classifier with attribute filtering.