Comparative Preferences Induction Methods for Conversational Recommenders

Abstract

Abstract. In an era of overwhelming choices, recommender systems aim at recommending the most suitable items to the user. Preference handling is one of the core issues in the design of recommender systems and so it is important for them to catch and model the user's preferences as accurately as possible. In previous work, comparative preferences-based patterns were developed to handle preferences deduced by the system. These patterns assume there are only two values for each feature. However, real-world features can be multi-valued. In this paper, we develop preference induction methods which aim at capturing several preference nuances from the user feedback when features have more than two values. We prove the efficiency of the proposed methods through an experimental study.

Evaluating conversational recommender systems

“…Choice set reduction Some systems aim to reduce the number of remaining options during the conversation to increase efficiency. Related terms are the number of unique cases presented, result set size, remaining items, number of cases to inspect or pruning rate, e.g., (Shimazu, 2002;Rafter and Smyth, 2005;Trabelsi et al, 2013;Smyth and McGinty, 2003)…”

“…Task completion times (or: session times) are commonly used in user studies to inform about the time needed until a recommendation is found. In some cases, computation and running times are reported for simulation experiments (Llorente and Guerrero, 2012;Trabelsi et al, 2013).…”