PROPOSED KNNXGBOOST CLASSIFICATION MODEL FOR BREAST CANCER PREDICTION USING BIG DATA ANALYSIS

Abstract

The breast cancer identification at early stage remains a difficult task for radiologist and researchers who deal with breast cancer identification process. The big data and image processing coupled together makes a remarkable attempt to solve addressed issues in identification process. The procedure followed in big data mining is much related to the process followed in data mining but differs in handling the huge voluminous confidential medical data. The process followed in the early stage of the research is to collect the breast cancer infected data from WBC, were the preprocessing and feature extraction process is carried out for improving the quality of the collected data as well as to increase the efficiency of the featured breast cancer data. The proposed KNN-XGBOOST classification algorithm is implemented further in segmentation and classification process. The procedure helps in classifying the two different classes segmented in feature identification process using proposed PCA-K-mean clustering technique earlier. The severity levels in breast cancer can be easily identified using the proposed KNN-XGBOOST classification algorithm with minimum time. The proposed classification technique is tested with similar classification techniques for finding the accuracy of identification process. The proposed KNN-XGBOOST performance is better than the other existing classification techniques with best time taken for implantation.