Guidelines for Systematic Review in Discussion Forums

Discussion forums like StackOverflow are an essential part of day-to-day modern software development. From small bugs and tutorials to high level discussions about software architecture and project management. These forums bring with them the point of view of practitioners and reflect tendencies from the past and from the present, in other words, they are a meeting point of practitioners to discuss various topics regarding software development. Despite being open for everyone to contribute, discussion forums do not lack credibility. The questions, answers, comments and even users are subject to community-moderation and reputation systems, which bring more confidence to data. Despite the popularity of discussion forums among both practitioners and researchers, the literature lacks approaches that guide the process of analyzing and systematically reviewing this type of literature. This becomes especially important given the massive amount of data that can be generated with queries to these datasets. In this context, this work aims to build a guideline and tools that can help researchers in the systematic review of gray literature from discussion forums. This approach enables a faster and more streamline process of extraction and analysis, which can help practitioners and researchers to obtain useful data and insights. We are currently performing studies in various forums and the results so far pointed out some aspects that can be used to categorize the forum posts.