The identify-assess-control cycle in goal-oriented requirement engineering aims at identifying, assessing, and resolving divergences in which the goals of the requirement cannot be satisfied as a whole. The boundary conditions (BCs) have shown great potential in requirements engineering because a BC captures a particular combination of circumstances, i.e., divergence. Existing researches have attempted to automatically identify lots of BCs. Unfortunately, a large number of identified BCs in the identification stage make the assessment stage and the resolution stage very expensive, and even impractical. We observe that existing identifying methods mainly adopt generality metric in order to filter out redundant BCs. However, a set of general BCs still retains a large number of redundant BCs since the generality metric can be considered as a coarse-grained metric to only filter out less general BCs, and a general BC potentially captures redundant circumstances that do not lead to a divergence. Furthermore, the accuracy of the assessment step based on likelihood is sensitive to redundant circumstances, so a set of general BCs leads to mistakes in the assessment step, which results in costly repeatedly assessing and resolving divergences.

In this paper, we present a novel metric to filter the redundant BCs, which is a fine-grained metric to improve the accuracy of the assessment step. We first introduce the concept of contrasty of BCs from the point of resolving divergences. Intuitively, if two BCs are contrastive, they capture the different divergences of the requirements specifications. We argue that a set of contrastive BCs should be recommended to requirements engineers, rather than a set of general BCs that potentially only indicates the same divergence. Then we design a post-processing framework (PPAc) to produce a set of contrastive BCs. Experimental results show the contrasty metric can filter out all the BCs that capture the same divergence, which dramatically reduces the number of BCs recommended to engineers. Furthermore, experiments also show that the BCs identified by the start-of-the-art method are not contrastive in most cases, which means that the BCs capture the same divergence, which makes the identify-assess-control cycle inefficient. In order to improve efficiency, we propose a joint framework (JAc) to interleave assessing based on the contrasty metric with identifying BCs. The primary intuition behind JAc is that it considers the search bias towards the BCs that capture different divergences. Experiments show that JAc produces the search bias towards contrastive BCs and identifies contrastive BCs more efficiently than PPAc.