A Comparative Study of Compound Critique Generation in Conversational Recommender Systems

摘要: Critiquing techniques provide an easy way for users to feedback their preferences over one or several attributes of the products in a conversational recommender system. While unit critiques only allow critique attribute each time, well-generated set compound enables input on at same and can potentially shorten interaction cycles finding target products. As result, dynamic generation is critical issue designing critique-based systems. In earlier research Apriori algorithm has been adopted generate from given data set. this paper we propose alternative approach generating based multi-attribute utility theory (MAUT). Our automatically updates weights product as result interactive critiquing process. This modification then used determine according those with highest values. experiments show that generated by are more efficient helping find than algorithm.