%0 Conference Paper %1 brusilovsky:MahmoodRicciHT09 %A Mahmood, Tariq %A Ricci, Francesco %B Proceedings of the 20th ACM conference on Hypertext and hypermedia %C New York, NY, USA %D 2009 %I ACM %K recommender %P 73--82 %R 10.1145/1557914.1557930 %T Improving recommender systems with adaptive conversational strategies %U http://dx.doi.org/10.1145/1557914.1557930 %X 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. %@ 978-1-60558-486-7

@inproceedings{brusilovsky:MahmoodRicciHT09, 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.}}, added-at = {2018-03-19T12:24:51.000+0100}, address = {New York, NY, USA}, author = {Mahmood, Tariq and Ricci, Francesco}, biburl = {https://www.bibsonomy.org/bibtex/209f59f9da4949dd68dc0c7c3f8fb3e5b/aho}, booktitle = {Proceedings of the 20th ACM conference on Hypertext and hypermedia}, citeulike-article-id = {5023627}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1557914.1557930}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1557914.1557930}, comment = {(private-note)Conversational recommender, which uses markov decision process to decide which out of many questions to ask taking many parameters. The target is to minimize reward. The system plan in advance insread of trying to choose the best step in the current moment}, doi = {10.1145/1557914.1557930}, interhash = {8b66c6c4995ed720d1b6b0029cbb36c9}, intrahash = {09f59f9da4949dd68dc0c7c3f8fb3e5b}, isbn = {978-1-60558-486-7}, keywords = {recommender}, location = {Torino, Italy}, pages = {73--82}, posted-at = {2009-06-30 16:39:19}, priority = {2}, publisher = {ACM}, series = {HT '09}, timestamp = {2018-03-19T12:24:51.000+0100}, title = {{Improving recommender systems with adaptive conversational strategies}}, url = {http://dx.doi.org/10.1145/1557914.1557930}, year = 2009 }