Performance Analysis of Mammogram Tumor Classification using Deep Belief Network

Abstract

Aim: The main aim of the research is to analyze theperformance analysis of mammogram tumor image classification using Deep Belief Network (DBN) over Decision Tree (DT) with improved accuracy. Materials and Methods: The research includes two groups namely Decision Tree (DT) as group 1 and Deep Belief Network (DBN) is considered as group 2 algorithms are used here to find the accuracy of mammograms. Each group consists of 25 samples with a total sample size of 50 to evaluate the accuracy. For statistical analysis the SPSS tool was used. The sample size was calculated using G power with pretest power at 80%. Result: The accuracy of DBN is significantly improved with percentage and there is a statistical significance observed as 0.015 (p < 0.05). The mean accuracy and standard deviation for Group 1 is 88.54% with 0.60 and for group 2 is 94.52% with 0.89. Conclusion: The NovelDeep Belief Network (DBN) algorithm is significantly accurate compared to the Decision Tree (DT) to analyse the performance analysis of mammogram tumor image classification.