Stance Detection

Stance detection is a recent and significant research problem in natural language processing which, in turn, is considered a subproblem of artificial intelligence. It is defined as the task of determining the stance of the author of a piece of text towards a target from the corresponding text. Commonly used classes of stance detection are Favor, Against, and Neither. Stance detection, particularly on microblogs texts, can be used to facilitate predictions for elections, polling/surveys, and trend analysis, among others. In this talk, we will cover basic concepts regarding stance detection, in addition to descriptions related research problems, approaches to stance detection, its application areas, and outstanding issues.

Dilek Kucuk

Dilek Kucuk is a chief researcher and associate professor at the Energy Institute of TUBITAK Marmara Research Center (MRC) in Ankara, Turkey. She is also the leader of Power Systems Information Technologies Group at the institute. Her group is the recipient of the best research group award of TUBITAK MRC for the year 2017. She received her B.Sc., M.Sc. and Ph.D. degrees in Computer Engineering all from Middle East Technical University in 2003, 2005, and 2011, respectively. Between May 2013 and May 2014, she studied as a post-doctoral researcher at European Commission's Joint Research Centre in Italy. Her research interests include stance detection, social media analysis, natural language processing, ontologies, energy informatics, and data mining. She is the author or co-author of 16 papers published in SCI-indexed journals and 37 papers presented at international conferences/workshops.