Automation of Software System Development Using Natural Language Processing and Two-Level Grammar

摘要

In software engineering, even with recent active research on formal methods and automated tools, users' involvement is inevitable and crucial throughout the software development lifecycle. Automation of these manual tasks would assist the developers throughout the development. Our project goal is to help the engineers to resolve ambiguity in natural language (NL) using Natural Language Processing and to overcome different levels of abstraction between requirements documents and formal specifications using Two-Level Grammar (TLG). The result is a system that assists developers to build a formal representation from the informal requirements for rapid prototyping and complete system implementation.