The acquisition, analysis and evaluation of imprecise requirements for knowledge-based systems

In this paper, a theoretical foundation has been laid and a practical method has been developed for specifying, analyzing and evaluating the complex relationships between imprecise requirements in knowledge-based systems. Imprecise requirements are represented by the canonical form in test-score semantics. The relationships between requirements are classified to be conflicting and cooperative based on the qualitative and quantitative analysis of relationships between requirements. This kind of analysis makes it possible to formulate a feasible overall requirement from conflicting individual requirements. It also facilitates to find better trade-off strategies for conflicting requirements by using fuzzy multi-criteria optimization technique. A requirement engineering process has also been developed to incorporate imprecise requirements into the requirement analysis for knowledge-based systems.