Abstract: Problems in software engineering area can be solved mathematically. In this study, Support Vector Machine (SVM) are utilized to categorize Non-Functional Requirements (NFRs). The NFR-Classifiers are used to identify cross-cutting predominant framework toward disintegration found in necessities particular or early plan records are proposed. Optimization is used to acquire the best results under given circumstances. In order to improve the efficiency of SVM, Artificial Bee Colony (ABC) technique with Differential Evolution (DE) is used. The proposed technique improves the classification accuracy by 90.54% than existing techniques.
K. Mahalakshmi, S. Manikandan, S. Nithyanantham, K.A. Sathiyaseelan and P. Sudhakar, 2016. Unsupervised Learning Technique Using Hybrid Optimization for Non-Functional Requirements Classification. Asian Journal of Information Technology, 15: 1457-1467.