2020/09/03 by Gustavo Polleti, Polleti, Gustavo Padilha, Douglas Luan de Souza +3
Computer Science · #68T01 #Advanced Graph Neural Networks #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #I.2.1 #Information Retrieval (cs.IR) #Recommender Systems and Techniques #Social and Information Networks (cs.SI) #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2009.01953
openalex publication_date 2020/09/03 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
A few Recommender Systems (RS) resort to explanations so as to enhance trust\nin recommendations. However, current techniques for explanation generation tend\nto strongly uphold the recommended products instead of presenting both reasons\nfor and reasons against them. We argue that an RS can better enhance overall\ntrust and transparency by frankly displaying both kinds of reasons to users.We\nhave developed such an RS by exploiting knowledge graphs and by applying\nSnedegar's theory of practical reasoning. We show that our implemented RS has\nexcellent performance and we report on an experiment with human subjects that\nshows the value of presenting both reasons for and against, with significant\nimprovements in trust, engagement, and persuasion.\n