The A* International World Wide Web Conference 2017 (WWW) is an annual forum that brings together the most influential researchers of the Web community. Thus, it is a special honour for Dominik Kowald and Elisabeth Lex from the Social Computing area of the Know-Center that their paper “Temporal Effects on Hashtag Reuse in Twitter: A Cognitive-Inspired Hashtag Recommendation approach” was accepted for this venue. This year, only 164 out of 966 valid submissions were accepted, which results in a very competitive acceptance rate of only 17%. The conference will be held from April 3 – 7, 2017 in Perth, Western Australia.
The paper analyses how the effect of time influences the use of hashtags in Twitter in order to propose a novel tag recommendation algorithm based on a model of cognitive science. In this respect, the authors find that this novel approach is able to outperform other related algorithms from the research area of tag recommendations with respect to recommendation accuracy and ranking. The underlying cognitive model used in this paper has already been exploited for recommender systems in course of the Learning Layers project and is now further investigated in the AFEL project.
A pre-print of the paper can be downloaded here.
Full bibliography: Kowald, D., Pujari, S., & Lex, E. (2017). Temporal Effects on Hashtag Reuse in Twitter: A Cognitive-Inspired Hashtag Recommendation Approach. In Proceedings of the 26th International World Wide Web Conference (WWW’2017). ACM.