A new efficient learning approach E-PDLA in assessing the knowledge of breast cancer dataset

Breast cancer is a deadly cancer that develops from the breast tissue. It is one among few reasons for women deaths in the world. Data mining and modern data analytics methods provide excellent support to infer knowledge from the existing database. Application area like medical world needs such efficient automated knowledge inferring tools and methods for better decision making. In this work, efficient data analytic methods like, K-nearest neighbours (KNN), C4.5 decision tree, naïve Bayes (NB), support vector machine (SVM), expert-pattern driven learning architecture (E-PDLA) are performed against the Wisconsin Breast Cancer (WBC) dataset. The aim of the work is to diagnose the proficiency and operational capabilities of the algorithms that have been tested. The results have been tabulated with the possible performance metrics and found the E-PDLA gives highest accuracy (99%) on classifying the dataset which gives insight of knowledge. All experiments have been simulated in J2EE environment with support of weka tool.