Qualitative Analysis Model for Software Requirements Driven by Interviews

Despite the recent precision of Software Engineering (SE), software systems still suffer from foremost problems, which are constantly referenced to Requirements Engineering (RE). RE is both the hardest and most critical part of software development, since errors at RE stage postponement to the end of the software development process and the hardest to repair. One of the major problems in RE is the analysis and interpretation of data driven by interviews, in view of the fact that most data comes in narrative, text, story and informal shape, which is generally called qualitative data. As well as data comes in mass appearance, which is later difficult to interpret to software requirements. In this study we have offered a brief overview of RE, interviews and qualitative models. Originally, we have offer a key improvement in the analysis of interviews, by proposing a Bottom-Up Qualitative Analysis Model (BUQAM) for enhancing the analysis of data driven by interviews, consequently well interpretation to software requirements. The BUQAM consist of the following steps: (1) Focused Oriented Questions (2) Organizing and Identifying Data (Familiarization) (3) Iterative Coding and Pattern Categorizing (4) Requirements Mapping and Classification and (5) Requirements Generation. We argue that this model will renovate most of narrative (descriptive) data into interpretive (clear-cut) information. In the same time, this model can furthermore be used in several types of narrative data such as ethnographic studies, action videos, observations, note files and surveys. In order to clarify our assumptions, we have validated this method against a case study for developing a University Wireless Network System (UWNS).