A method to support search string building in systematic literature reviews through visual text mining

Despite the increasing popularity of systematic literature reviews in Software Engineering, several researchers still indicate it as a costly and challenging process. Aiming at alleviating this costly process, we propose an iterative method to support the process of building the search string for a systematic review. This method uses Visual Text Mining techniques to support the researcher by suggesting new terms for the string. In order to do so, the method extracts relevant terms from studies selected by the researcher and displays them in a way that facilitate their visualization and supports building and refining the search string. In order to check the feasibility of this approach, we developed a tool that implements the proposed method. Interviews with researchers identified their difficulties in performing systematic reviews and captured their feedback with regards the use of the proposed method in a user study. The researchers indicated that this approach could be used to improve the process of building the search strings for systematic reviews. The study indicates that our approach can be used to facilitate the construction of the systematic literature review search string.