Article Preview
Top1. Introduction
Ascertaining user requirements is the foremost exercise in any product, more so Business-to-Business (B2B) product development. The paper pertains to mining the user-requirements for software development.
The paper acknowledges that the user requirements, especially in multi-use, multi-domain, multi-location environments that most large businesses of today are characterised by, are multiple, diverse and variously expressed. This necessitates development of a methodology that permits effective sourcing, coalescing and collating of the user-requirements.
The paper proposes Apriori alogrithm as the preferred technique for mining the user requirements. It addresses and automates two prime issues of user requirements in an integrated manner. One, it focuses on knowledge-driven elicitation of user requirements, where knowledge is extracted in the form of mutual association and frequent occurrence of user-requirements. Two, it stratifies the elicited user-requirements into two categories-“Top Priority Apriori List” and “Low Priority Apriori List”- that reflect the comparative strength of the concerns of all categories of users and prime stakeholders in the development of the software. Thus, Apriori algorithm filters out the mundane details in the user descriptions of their requirements (Mussabacher, 2016; Wong, Mauricio, & Rodriguez, 2017; AlMousa, Al-Khalifa, & AlSobayel, 2017). Further, the requirements so elicited are validated with the help of a fuzzy linguistic survey to generate “High Priority Survey List” and “Low Priority Survey List.” The paper uses two-way gap analysis to ensure the correctness and completeness of the user requirements.