Beyond the Systematic: Forecasting Importance and Emergence of Research Areas in Applications of Software Traceability Using NLP

Secondary studies, such as systematic literature reviews (SLR) and systematic mapping studies (SMS), are commonly conducted in any discipline to search, appraise and collate all relevant empirical evidence in order to provide a complete interpretation of research results to answer research questions. These also help to provide insight into future research work, which is typically achieved by observing gaps in past works and hinting at the possibility of future research in those gaps. By combining NLP and time series, we propose a meta-analysis to extend current systematic methodologies of literature reviews and mapping studies. Our work uses a Word2Vec model, pre-trained in the software engineering domain, and is combined with an autoregressive integrated moving average (ARIMA) time series model. Our aim is to forecast future trajectories of research outlined in systematic studies, rather than just describing them, specifically in the field of software traceability. Using the same dataset from our own previous mapping study, we were able to go beyond descriptively analysing the data that we gathered. In this paper, we continue and expand on our previous work to show the emerging importance of terms linked to requirements and design in software traceability and the forecast for the following periods based on ARIMA time series modelling. Our proposed methodology is an exemplar for exploring the potential of language models coupled with time series in the context of systematically reviewing the literature.