Machine Learning Promising Prediction in Feature Extraction

Abstract

The world is crunching with high volumes data and high end technologies, instead our still news edition giving place to �cancer affected died for the cause of no recognition but not merely due to cancer�. This paper provides the identification, feature extraction of cancer on the board of machine learning. The achievement of prediction accuracy rate improvement through the defined algorithms namely, KNN, Fast KNN,GLM,SVM is done. Initial way of performing the cross fold validation and checking the fitted turn out of feature extraction is the major contribution with 569 object data set and 32 attribute value of breast cancer data. Secondly GLM-Net with Feature Extraction using KNN and comparisons with SVM classifier for feature extraction with KNN, and subsequently SVM Basic model with Radial function without feature extraction is achieved, Finally SVM Classifier with Radial Basis function on Breast Cancer Dataset and Regularized Linear Support Vector Machines with Class Weights through RFE done and concluded as Fast KNN and SVM are the most promising classifiers of Machine learning and futuristic data science classification evaluators.