Building Domain Familiarity for Elicitation: A Conceptual Framework Based on NLP Techniques

Abstract

Elicitation is the entry point of any digital transformation project. Domain knowledge is a critical factor for successful elicitation. Hence, elicitation is usually executed by subject matter experts (SME), who have fair amount of domain knowledge required for the project. However, such SME are scare and costly resources. Hence there is an urgent need to design a framework that can help build domain knowledge of the business analyst (BA) involved in elicitation in order to lower dependence on SME. Such framework should be programmatically implementable and should be able to work seamlessly with the existing techniques of elicitation. Several solutions based on organizational theories, cognitive models, and strategic frameworks have been proposed with varying results. This work is in progress. As elicitation generates a web of interactions in natural language, natural language processing (NLP) techniques can be utilized to extract and build domain knowledge. This paper investigates the application of NLP to extract working domain knowledge from the available organizational documents to assist analyst during elicitation phase. The paper proposes a conceptual framework to extract and build domain familiarity. This paper also analyzes the gaps and new areas of research in this direction.