![]() | Only 14 pages are availabe for public view |
Abstract Breast cancer is one of the serious diseases that threaten the lives of many women worldwide in general and in Egypt in particular, where breast cancer causes the death of thousands of Egyptian women annually. The seriousness of this disease is that it is often discovered in late stages after a period of its occurrence, which causes a wide spread of the disease and difficulty in its treatment. Another important characteristic of this disease is that it is a disease that can return again after a period of treatment. Therefore, predicting the occurrence or recurrence of such disease early is the best solution to have a high cure rate. Therefore, the main objective of this thesis is to improve the prediction performance of breast cancer recurrence. Many previous methods have been proposed to predict breast cancer recurrence. The highest accuracy achieved from the previously proposed methods is 89.89% using one of the most famous datasets in the field of breast cancer recurrence’s prediction (i.e., Wisconsin Prognosis Breast Cancer (WPBC) dataset). This thesis provides a framework for improving the prediction of breast cancer recurrence. The proposed framework has the ability to overcome many of the challenges in the existing dataset such as the problem of imbalance between the classes, and the large number of data dimensions. It also uses the neural network algorithm to fuse the results of a number of individual classifiers. Our proposed framework evaluation showed a significant improvement in predication performance. It achieved an accuracy of 9.3%, area under the curve of 99%, and precision, recall, and f1-measure of 98%. |