Método para apoio à construção de strings de busca em revisões sistemáticas por meio de mineração visual de texto

Abstract

Despite the increased popularity of the adoption of Systematic Literature Reviews in Software Engineering, many researchers still indicate it as a costly and challenging process. Studies report problems in different activities throughout the review process, as in the construction of the Systematic Review search string and selection of primary studies. Aiming to promote aid to its realization, tools based on methods and techniques from the Visual Text Mining area are presented in published studies, proposing assistance in various tasks of a Systematic Literature Review. However, it s perceived a lack of methods proposing to aid a researcher with the construction of the Systematic Review search string, on its planning phase. In this context, this paper proposes an iterative method to assist the process of building the search string for a Systematic Review. Using Visual Text Mining techniques, it supports the researcher by suggesting terms for the search string. Relevant terms are extracted from studies selected by the researcher and shown in a visualization that facilitates the decision of the researcher to update the search string and include them, building and refining the search string that will be used in the Systematic Review. A tool that implements the proposed method has been developed, allowing the execution of tests with researchers and an analysis of the feasibility of this proposal. Interviews with researchers identified the difficulties in performing Systematic Reviews and captured their opinions regarding the use of the proposed method, discussing its adoption