| Requirements analysis is knowledge intensive and requires an expert to understand what the user needs. In this research, not only the ordinary functional requirements are collected, but also the non-traditional information which is the non-functional requirements like the quality of operations or the background information for constructing the requirements are gathered through a knowledge base support. Different kinds of information collected are stored and organized in a knowledge base and can be used as the source of user input in the latter phases of software development.;This research is a result of applying techniques from the artificial intelligence field to software engineering problems. With a knowledge-based support, it offers a systematic way toward requirements analysis, specification production, prototype generation, specification debugging, and code transformation.;Algorithms and procedures are provided for constructing interface language to the user, organizing the knowledge base, and applying the knowledge base to different tasks. A non-monotonic reasoning technique is used for handling the specification and the update of the knowledge base. A prototype system, called Requirements Analysis and Knowledge Elicitation System (RAKES), is introduced as an example to describe this approach. Algorithms for knowledge-based specification debugging are also developed. A guide to apply this approach to other software engineering discipline such as transformation and quality assurance is also discussed in detail. |
