COMPARATIVE ANALYSIS ON MACHINE LEARNING CLASSIFICATION APPROACHES FOR EFFICIENTBUILDING SOFTWARE REQUIREMENTS

Software Requirements are the premise of top notch software advancement process, each progression is identified with SR, and these speak to the necessities and desires for the software in an itemized structure. The software requirement classification (SRC) task requires a great deal of human exertion, extraordinarily when there are enormous of requirements, in this way, the mechanization of SRC have been tended to utilizing Natural Language Processing (NLP) and Information Retrieval (IR) systems, notwithstanding, for the most part requires human exertion to break down and make highlights from corpus (set of requirements). In this work, the model that we propose depends on to create code assessment capable framework utilizing AI calculation in software building.1. To give strong code audit capable framework utilizing SVM, KNN and Decision Tree classification. 2. To assess the presentation of the proposed method utilizing precision, accuracy, sensitivity and specificity parameters.