Improving recommender systems with adaptive conversational strategies

Abstract

Conversational recommender systems (CRSs) assist online users in their information-seeking and decision making tasks by supporting an interactive process. Although these processes could be rather diverse, CRSs typically follow a fixed strategy, e.g., based on critiquing or on iterative query reformulation. In a previous paper, we proposed a novel recommendation model that allows conversational systems to autonomously improve a fixed strategy and eventually learn a better one using reinforcement learning techniques. This strategy is optimal for the given model of the interaction and it is adapted to the users' behaviors. In this paper we validate our approach in an online CRS by means of a user study involving several hundreds of testers. We show that the optimal strategy is different from the fixed one, and supports more effective and efficient interaction sessions.

CAPRE: A New Methodology for Product Recommendation Based on Customer Actionability and Profitability

“…Then, section III discusses our methodology CAPRE with the two measures of actionability and profitability. We apply CAPRE in order to make recommendations on an actual dataset of VM Matériaux trading group 1 in section IV. We analyze the effectiveness of CAPRE in section V by performing a cross-validation on the MovieLens benchmark.…”

“…The dataset was converted into a binary usermovie matrix C × P that has 943 customers and 1,682 movies that were rated by at least one of the users. We apply our methodology on the five usual learning datasets named U [1][2][3][4][5].base and the five usual application datasets named U [1][2][3][4][5].test. Training and validation datasets contain 80 % and 20 % of all ratings respectively.…”

“…N ∪ I, are the age, the gender and the occupation. We gathered the ratings into three classes a = [1,2], b = [3] and c = [4,5].…”

“…RS are software tools and techniques providing suggestions for products to be use by a customer [1] [2]. They connect customers with potential products to purchase by associating either the content of products or the opinions of customers, or both.…”

Tourism Recommender Systems: An Overview of Recommendation Approaches

“…Recommender Systems (RS) are computer based tools and techniques providing suggestions for items to be of use to a user [1] [2] [3]. They have become an important research field since the emergence of the first paper on collaborative filtering in the mid-1990s [4] [5] [6] [7].…”