Using Visual Text Mining to Support the Study Selection Activity in Systematic Literature Reviews
This work provides a potentially beneficial tool for researchers conducting Systematic Literature Reviews, aiming to improve the efficiency and accuracy of the study selection phase. It is an incremental improvement to an existing methodology.
This paper addresses the time-consuming and manual process of primary study selection in Systematic Literature Reviews (SLRs) by proposing a novel approach called SLR-VTM, which utilizes visual text mining (VTM) techniques. A case study with doctoral students demonstrated that using SLR-VTM reduced the time spent on study selection and increased the number of correctly included studies.
Background: A systematic literature review (SLR) is a methodology used to aggregate all relevant existing evidence to answer a research question of interest. Although crucial, the process used to select primary studies can be arduous, time consuming, and must often be conducted manually. Objective: We propose a novel approach, known as 'Systematic Literature Review based on Visual Text Mining' or simply SLR-VTM, to support the primary study selection activity using visual text mining (VTM) techniques. Method: We conducted a case study to compare the performance and effectiveness of four doctoral students in selecting primary studies manually and using the SLR-VTM approach. To enable the comparison, we also developed a VTM tool that implemented our approach. We hypothesized that students using SLR-VTM would present improved selection performance and effectiveness. Results: Our results show that incorporating VTM in the SLR study selection activity reduced the time spent in this activity and also increased the number of studies correctly included. Conclusions: Our pilot case study presents promising results suggesting that the use of VTM may indeed be beneficial during the study selection activity when performing an SLR.