Structuring a Folksonomy in a Community of Questions and Answers

Abstract

Knowledge communities are composed of individuals who share the same interests and voluntarily work together to expand knowledge and understanding of a domain through learning and sharing. When a user asks a question in a knowledge community, he or she may usually add tags to categorize the question. These tags make it easy for experts to find the question and then answer it. One of the problems with this tagging system is that there is no hierarchy to aid in the search for broader themes. This study proposes a way of hierarchizing the folksonomy of an online knowledge community. In order to verify the applicability of the proposed approach, we used the Stack Exchange’s biology community. Through a crowdsourcing evaluation it was possible to show that the developed approach performed better than other heuristics and it was also possible to identify quality issues on the description of WordNet synsets.