REQUIREMENTS PRIORITIZATION USING NEURAL NETWORKS AND DEEP LEARNING

Abstract

Requirements prioritization is essential for effective software development but remains challenging due to its subjective and complex nature. This study presents a deep learning–based approach using neural networks to automate and enhance the prioritization of software requirements. By integrating natural language processing with contextual and sentiment-based features, the proposed model learns meaningful patterns from both structured and unstructured data. Experimental results show that the model outperforms traditional and machine learning methods in both classification accuracy and ranking effectiveness. The findings highlight the potential of deep learning to deliver scalable, consistent, and data-driven prioritization in modern software engineering environments.