What’s the problem? How crowdsourcing and text‐mining may contribute to the understanding of unprecedented problems such as COVID‐19

Abstract

In this research, we explore how crowdsourcing combined with text-mining can help to build a sound understanding of unstructured, complex and ill-defined problems. Therefore, we gathered 101 problem descriptions contributed to a crowdsourcing contest about the impact of COVID-19 on the tourism industry. Based on our findings we propose a five-phase process model for problem understanding consisting of: (1) information gathering, (2) information pre-structuring, (3) problem space mapping, (4) problem space exploration, and (5) problem understanding for solution search. While our study confirms that crowdsourcing and text-mining facilitate fast generation and exploration of problem spaces at limited cost, it also reveals the necessity to follow certain process steps and to deal with challenges such as information loss and human interpretation. For practitioners, our model presents a guideline for how to get a faster grasp on complex and rather unprecedented problems.