Soluções computacionais para suporte à condução de revisões sistemáticas da literatura em engenharia de software: um mapeamento sistemático

Abstract

Context: The academic community in the field of Software Engineering (SE) has increasingly re- lied on Systematic Reviews (SRS), such as Systematic Literature Reviews (SLRs) and Systematic Mappings (SMs), due to the numerous benefits they provide, including greater methodological rigor, reproducibility, and support for identifying research gaps. However, computational support for the SR process still presents significant limitations, highlighting the need for the development of solutions capable of reducing the time and effort required for their execution. Objective: This study aims to identify and analyze the main computational solutions developed to support the conduction of SRs in Software Engineering. Method: A systematic mapping of the literature was conducted, covering studies published between 2007 and August 2025, with the goal of answe- ring four research questions. Results: The study identified 45 computational approaches that provide support for SLRs in SE, most of which focus on the conduction phase and are implemented as tools or algorithms. Despite the recent increase in research activity, many solutions remain in early stages of development, showing limitations in terms of usability and integration. The main challenges observed include bias in automation, the construction of search strategies, data management, and the lack of integrative tools. Conclusion: The results highlight the need for more cohesive and comprehensive computational support across all stages of SLRs.