A CASE-BASED REASONING SYSTEM FOR INTELLIGENT STRUCTURED SOFTWARE DEVELOPMENT

THE OBJECTIVE IS TO DEVELOP A KNOWLEDGE-BASED ASSISTANT FOR THE SOFTWARE ENGINEERING (SE) PROCESS. A CASE-BASED-REASONING (CBR) APPROACH TO KNOWLEDGE ELICITATION IS PROPOSED AS A SENSITIVE AND EFFECTIVE STRATEGY TO DERIVE AND REPRESENT CRITICAL KNOWLEDGE. PHASE I WILL DEMONSTRATE THE FEASIBILITY OF USING A CBR APPROACH TO DEVELOPING A KNOWLEDGE-BASED ASSISTANT FOR THE EXTENSIVE AND COMPLEX TASKS ASSOCIATED WITH SOFTWARE ENGINEERING EFFORTS. CASE TOOLS HAVE BEEN BUILT TO ASSIST IN A PORTION OF THE SOFTWARE ENGINEERING PROCESS, BUT HAVE BEEN LIMITED IN RESULTS BY THEIR VIEW OF THE REQUIREMENTS NEEDED FOR DOMAIN SPECIFIC PROBLEMS. A KNOWLEDGE-BASED ASSISTANT IS REQUIRED TO PROVIDE THE EXPERIENCE OF SES IN A STRUCTURED, AND ENCAPSULATED FORMAT. THROUGH PROPER ELICITATION TECHNIQUES A CASE-BASE OF PRIOR EXPERIENCES CAN BE DEVELOPED TO PRODUCE A CBR SYSTEM. THIS SYSTEM CAN PROVIDE A CLASSIFICATION STRUCTURE FOR BOTH THE FORMAL AND INFORMAL METHODS USED BY SES TO DESIGN, IMPLEMENT, AND TEST SOFTWARE DESIGNS.