A preliminary study on the suitability of stack overflow for open innovation in requirements engineering
Huishi Yin, Dietmar Pfahl · Proceedings of the 3rd International Conference on Communication and Information Processing · 2017
Due to lack of resources (funds, social influence, user feedback etc.), small companies, in particular, start-up companies, almost exclusively must rely on their own expertise with regards to technological innovation. Therefore, it could be beneficial for them to analyze big companies or competitors' products in order to get a better understanding of what users think about them. The core idea of my research project is to use opinions voiced by users about strengths and weaknesses of competitors' products to improve the understanding of actual user needs and thus help focus the requirements definition of the small, especially start-up company's own product. In order to lower the cost of analysis, we investigate how user needs can be distilled from online open sources. Since there are different types of online open sources, as a starting point of our current stage of research, we focused on one type of online open sources - question and answer (Q&A) sites such as Stack Overflow. As a showcase, we used the IDE project Eclipse. In our exploratory study, we analyzed the popularity of Eclipse, the amount of questions asked, the amount of questions asked that received an accepted answer, as well as the contents of questions asked about Eclipse. We also analyzed the emotional polarity of the posted questions. We found that the Stack Overflow timely updates, and the emotions reflected is more objective according to the result shown by sentiment analysis that 82% questions show neutral emotion.