A Novel Hybrid System for Diagnosing Breast Cancer Using Fuzzy Rough Set and LS-SVM

R. Jaya Suji ； S. P. Rajagopalan

《Research Journal of Applied Sciences, Engineering and Technology》 10卷1期 (2015/05) Pp. 49-55

並列摘要

With fast development of medical diagnosis technologies, the filtering of the entire relevant feature and time consuming task are challenging tasks. For effective feature selection and reducing the time consuming, we propose a new hybrid system for diagnosing the breast cancer. The proposed hybrid system is the combination of CFRSFS, K-Means Clustering and Least Square Support Vector Machine (LS-SVM). In this hybrid system, we propose a new feature selection algorithm called Correlation based Fuzzy Rough Set Feature Selection (CFRSFS) algorithm for effective initial feature selection process. Moreover, K-Means clustering algorithm has been used for enhancing the feature selection process based on the factors of the selected features in similar manner of the existing hybrid system. Finally, LS-SVM algorithm is also used for classifying the feature selected breast cancer dataset. The experiments have been conducted for evaluating the proposed system using WDBC Data set. The obtained results show that the performance of the proposed system classification accuracy is 99.54%.

並列關鍵字

Cancer diagnosis ； data mining ； k-means clustering ； least square support vector machine