Application of machine learning and natural language processing in automated requirement analysis and object-oriented system design for adaptive software development environments

The translation of natural language software requirements into formal design artifacts remains one of the most labor-intensive and error-prone phases of the software development lifecycle (SDLC). Ambiguity, inconsistency, and the rapid evolution of project requirements often lead to misaligned system architectures. This research investigates the application of Machine Learning (ML) and Natural Language Processing (NLP) techniques to automate the extraction of entities and relationships from requirement specifications to generate Object-Oriented (OO) design models. By implementing an adaptive framework, this study proposes a system capable of continuous refinement as requirements change. The findings demonstrate a significant reduction in architectural modeling time and improved alignment between stakeholders’ natural language specifications and the final technical design.