ACQUIRING APPLICATION-SPECIFIC KNOWLEDGE DURING DESIGN TO SUPPORT SYSTEMS MAINTENANCE

摘要

Most large systems development efforts proceed in a top-down fashion where initialspecifications and requirements are incorporated into a high-level design, followed byprograms based on this design. However, a major part of the software life-cycle effort isdevoted to maintenance. While several existing methodologies aid in the initial phases ofrequirements and specification, they have proven to be of little value for maintenance.Changes in user requirements are often translated directly to the level of code, divorcingit from the high level design it was based on. After a few such changes, the programsmay not correspond to any formal high-level design, making subsequent maintenancedifficult. We argue that maintenance must be based on the knowledge used insynthesizing the high-level design. This requires a development environment where theknowledge about high-level designs is formally represented, and raises the questionabout how this knowledge will be acquired by the support environment in the firstplace. In this paper, we present a model that enables the support environment to acquiredesign knowledge through "learning by observation" of a designer engaged in specifyinga high-level design. The knowledge that the learning system begins with is a genericobject for expressing design decisions. Based on the input provided by the designer, anda limited interactive querying process, it constructs and continuously refines ataxonomic classification of application-specific knowledge and rules at an appropriatelevel of generality that capture the rationale of the design. This knowledge can be usedsubsequently for maintaining system designs and recognizing design situations similar tothe ones it has knowledge about.