부산대학교 도서관

In recent years, the number of breast cancer patients has increased rapidly. To increase the survival rate, the early diagnosis of breast cancer is very important. Hence, a trustworthy diagnosis and detection process is required. Automatic detection processes will be very helpful for medical practitioners. There are numerous proposed methods for timely sensing of breast cancer This study has suggested a deep neural network with feature selection techniques to predict breast cancer. The appraisal of the suggested strategy is performed on different evaluation benchmarks like train accuracy, test accuracy, precision, recall, specificity, sensitivity, f measure and MCC. Simulation result of the proposed method is very promising (accuracy 99.42%). Based on the findings of experimental simulations and analysis of the statistical data, the suggested method is efficient, beneficial and more accurate to predict breast cancer. As stated by the presented literature review, It is like setting an example among the prevailing machine learning strategies.