
Despite the importance of conducting systematic literature reviews (SLRs) for identifying the research gaps in software engineering (SE) research, SLRs are a complex, multi-stage, and time-consuming process if performed manually. Conducting an SLR in line with the guidelines and practice in the SE domain requires considerable effort and expertise. The objective of this SLR is to identify and classify text-mining techniques and tools that can help facilitate SLR activities. This study also investigates the adoption of text-mining (TM) techniques to support SLR in the SE domain. We performed a mixed search strategy to identify relevant studies published from January 1, 2004, to December 31, 2016. We shortlisted 32 papers into the final set of relevant studies published in the SE, medicine and social science disciplines. The majority of the text-mining techniques attempted to support the study selection stage. Only 12 out of the 14 studies in the SE domain applied text-mining techniques, focusing primarily on facilitating the search and study selection stages. By learning from the experience of applying TM techniques in clinical medicine and social science fields, we believe that SE researchers can adopt appropriate SLR automation strategies for use in the SE field.
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 43 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Top 10% | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Top 10% | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Top 10% |
