Handling Inconsistency In Distributed Software Requirements Specifications Based On Prioritized Merging

Abstract

Developing a desirable framework for handling inconsistencies in software requirements specifications is a challenging problem. It has been widely recognized that the relative priority of

Measuring the blame of each formula for inconsistent prioritized knowledge bases

“…The corresponding prioritization function P I : L → Pri I is a function that assigns a precise numerical valuation to each formula such that the bigger the priority valuation of a formula, the more preferred the formula is. For example, the necessity degree attached with each formula in possibilistic knowledge base [6] [19,21].…”

“…Generally, the relative priority of a prioritized knowledge base should depend on priority levels of formulas belonging to that knowledge base. However, for qualitative prioritization, it is often hard to use a single value to capture the relative priority of a knowledge base consisting of qualitatively prioritized formulas, since it is very subtle to replace each qualitative level with a numerical weight [21]. Correspondingly, the operator of comparison and integration should take the stratified structure of such a knowledge base into account.…”

“…Generally, it is often hard to use a single value to measure the significance of such knowledge bases, since it is very subtle to replace each qualitative level with a numerical weight. Moreover, the weights may not be explained intuitively in many cases [21]. For example, consider {a},∅,{¬a} , if we assign weight 9 to a (at a relatively high level) and 1 to ¬a (at a relatively low level), we do no think that the degree of importance of a is exactly 9 times of that of ¬a.…”

“…That is, the current proposals for measuring the blames of formulas for inconsistency of a knowledge base do not take the priority of each formula into account. However, as a description of the importance or reliability of the corresponding knowledge, the priority of each formula always plays an important role in inconsistency handling in artificial intelligence and many applications, such as belief revision [4], knowledge base merging [6,24] and inconsistency handling in requirements engineering [20,21]. It is natural to take the priority of each formula into account when identifying the blames of formulas for inconsistency of a prioritized knowledge base.…”

Preferential Multi-Context Systems

“…Many (if not all) real-world applications of sharing and reasoning knowledge are characterized by heterogeneous contexts, especially with the advent of the world wide web. Research in representing contexts and information flow between contexts has gained much attention recently in artificial intelligence [11,4,7,8,5,13] as well as in applications such as requirements engineering [10,15,14].…”