BREAST CANCER DETECTION USING ENSEMBLE CLASSIFICATION AND EXTENDED WEIGHTED VOTING METHOD

Abstract

Breast cancer disease increases the mortality rate of women. Early detection of cancer helps to increase the life span and decreases the mortality rate. The daubechies-8 wavelet filter, discrete wavelet transform entropies such as Shannon, Log energy and Sure entropy were applied to extract the textural features of mammogram image. The ensemble classifiers K.Nearest Neighbor, Bayes and Support Vector Machine were used to classify. The novel method of extended weighted voting method is applied to obtain the accurate result which produces the increased performance of 97.3% of results.