A Synthesis of Research on Requirements Analysis and Knowledge Acquisition Techniques

TLDR

Requirements analysis and knowledge acquisition share nearly identical processes, yet researchers in each field largely remain unaware of developments in the other. The article aims to merge awareness between RA and KA by comparing representative techniques across communication obstacles, control focus, and the nature of understanding. The authors compare representative RA and KA techniques, grouping them by elicitation mode and evaluating them on communication obstacles, control focus, and the nature of understanding. The comparison reveals substantial common ground between RA and KA and proposes future research on rigorous technique evaluation, domain‑elicitation matching, synergistic effects, new techniques, and generalized versus specialized approaches.

Abstract

Requirements analysis (RA) involves end users and systems analysts interacting in an effort to recognize and specify the data and information needed to develop an information system. In the design of expert systems, a similar process of eliciting information, in this case human knowledge, has been studied under the banner of knowledge acquisition (KA). When examined closely, many entities and processes involved in RA and KA are almost identical. However, researchers in each area are seemingly unaware of the developments in the other area. In order to facilitate a merged awareness of both research streams, this article compares representative RA and KA techniques, which are grouped, according to elicitation mode, on three dimensions: communication obstacles, a technique's focus of control, and the nature of the understanding gained from using the technique. This comparison demonstrates that these two research streams have many things in common and that researchers in one area can benefit from developments in the other area. Additionally, this analysis leads to several suggested research areas: (1) rigorous examinations of these techniques as they are used to overcome communication obstacles and to enrich understanding; (2) investigations into the seeming match between certain elicitation types and problem domain categories; (3) examinations into synergetic effects of elicitation techniques; (4) development of more techniques for eliciting information requirements to serve emerging needs; and (5) comparisons of the relative advantage of generalized versus specialized elicitation techniques.