EXPERIMENTAL ANALYSIS OF DESIGN CHOICES IN MULTIATTRIBUTE UTILITY COLLABORATIVE FILTERING
Abstract
Recommender systems have already been engaging multiple criteria for the production of recommendations. Such systems, referred to as multicriteria recommenders, demonstrated early the potential of applying Multi-Criteria Decision Making (MCDM) methods to facilitate recommendation in numerous application domains. On the other hand, systematic implementation and testing of multicriteria recommender systems in the context of real-life applications still remains rather limited. Previous studies dealing with the evaluation of recommender systems have outlined the importance of carrying out careful testing and parameterization of a recommender system, before it is actually deployed in a real setting. In this paper, the experimental analysis of several design options for three proposed multiattribute utility collaborative filtering algorithms is presented for a particular application context (recommendation of e-markets to online customers), under conditions similar to the ones expected during actual operation. The results of this study indicate that the performance of recommendation algorithms depends on the characteristics of the application context, as these are reflected on the properties of evaluations' data set. Therefore, it is judged important to experimentally analyze various design choices for multicriteria recommender systems, before their actual deployment.
References
- ACM Trans. Inform. Syst. 23(1), 103 (2005), DOI: 10.1145/1055709.1055714. Crossref, Web of Science, Google Scholar
- IEEE Trans. Knowl. Data Engin. 17(6), 734 (2005), DOI: 10.1109/TKDE.2005.99. Crossref, Web of Science, Google Scholar
- Mach. Learn. 6, 37 (1991), DOI: 10.1007/BF00153759. Web of Science, Google Scholar
- Commun. ACM 41(8), (1998), DOI: 10.1145/280324.280330. Google Scholar
- J. Electr. Commun. Res. 3(3), 114 (2002). Google Scholar
-
J. S. Breese , D. Heckerman and C. Kadie , Empirical analysis of predictive algorithms for collaborative filtering , Proc. 14th Conf. Uncertainty in Artificial Intelligence ( 1998 ) . Google Scholar - User Model & User Adapt. Inter. 12, 331 (2002). Crossref, Web of Science, Google Scholar
-
G. Carenini , User-specific decision-theoretic accuracy metrics for collaborative filtering , Proc. "Beyond Personalization" Workshop, Intelligent User Interfaces Conference (IUI'05) ( 2005 ) . Google Scholar - IEEE Trans. Inf. Th. 13(1), 21 (1967), DOI: 10.1109/TIT.1967.1053964. Crossref, Web of Science, Google Scholar
-
J. Delgado and N. Ishii , Memory-based weighted-majority prediction for recommender systems , Proc. ACM-SIGIR'99, Recommend. Syst. Workshop ( 1999 ) . Google Scholar - ACM Trans. Inform. Syst. 22(1), 143 (2004), DOI: 10.1145/963770.963776. Crossref, Web of Science, Google Scholar
- Commun. ACM 35(12), 61 (1992), DOI: 10.1145/138859.138867. Crossref, Web of Science, Google Scholar
- Electr. Commun. Res. Appl. 1, 314 (2002), DOI: 10.1016/S1567-4223(02)00023-6. Crossref, Google Scholar
- Artif. Intell. 146(2), 149 (2003), DOI: 10.1016/S0004-3702(03)00013-4. Crossref, Web of Science, Google Scholar
- Inform. Retr. 5, 287 (2002), DOI: 10.1023/A:1020443909834. Crossref, Web of Science, Google Scholar
- ACM Trans. Inform. Syst. 22(1), 5 (2004), DOI: 10.1145/963770.963772. Crossref, Web of Science, Google Scholar
- Eur. J. Oper. Res. 130, 233 (2001), DOI: 10.1016/S0377-2217(00)00035-7. Crossref, Web of Science, Google Scholar
-
R. L. Keeney , Value-focused Thinking: A Path to Creative Decisionmaking ( Harvard University Press , Cambridge, MA , 1992 ) . Google Scholar - ACM Trans. Inform. Syst. 22(1), 1 (2004), DOI: 10.1145/963770.963771. Crossref, Web of Science, Google Scholar
- Exp. Syst. Appl. 27, 665 (2004), DOI: 10.1016/j.eswa.2004.07.001. Crossref, Web of Science, Google Scholar
- Inform. Manag. 42, 387 (2005), DOI: 10.1016/j.im.2004.01.008. Crossref, Web of Science, Google Scholar
- Inform. Serv. Use 25(2), 95 (2005). Crossref, Google Scholar
- N. Manouselis and C. Costopoulou, Designing multiattribute utility algorithms for collaborative filtering algorithms, Technical Report, Informatics Laboratory, Agricultural University of Athens, TR 181 (2006) (available from the authors) . Google Scholar
- N. Manouselis and C. Costopoulou, Designing a Web-based testing tool for multicriteria recommender systems, Eng. Lett. Special Issue on Web Engineering (in press) . Google Scholar
-
J. Masthoff , Modeling the multiple people that are me , Proc. User Modeling 2003 ,Lecture Notes in Artificial Intelligent 2702 , eds.P. Brusilovsky , A. Corbett and F. de Rosis ( Springer , Berlin , 2003 ) . Google Scholar -
L. Maritza , A comparison of several predictive algorithms for collaborative filtering on multivalued ratings , Proc. 2004 ACM Symp. Applied Computing (SAC'04) ( 2004 ) , DOI: 10.1145/967900.968112 . Google Scholar - ACM Trans. Inform. Syst. 22(3), 437 (2004), DOI: 10.1145/1010614.1010618. Crossref, Web of Science, Google Scholar
M. Montaner , B. Lopez and J. L. de la Rosa , Evaluation of recommender systems through simulated users, Proc. 6th Int. Conf. Enterprise Information Systems (ICEIS'04)3 pp. 303–308. Google Scholar-
H. Nguyen and P. Haddawy , DIVA: applying decision theory to collaborative filtering , Proc. AAAI Workshop Recommend. Syst. ( 1998 ) . Google Scholar - Engin. Apps Art. Int. 18, 781 (2005), DOI: 10.1016/j.engappai.2005.06.010. Crossref, Web of Science, Google Scholar
- Inform. Interact. Intell. 1(1), 9 (2001). Web of Science, Google Scholar
-
B. Price and P. R. Messinger , Optimal recommendation sets: covering uncertainty over user preferences , Proc. Inform. Ann. Meet. Denver 2004 ( AAAI Press , 2005 ) . Google Scholar P. Resnick , GroupLens: an open architecture for collaborative filtering, Proc. ACM CSCW (1994) pp. 175–186, DOI: 10.1145/192844.192905. Google Scholar- Commun. ACM 40(3), 56 (1997), DOI: 10.1145/245108.245121. Crossref, Web of Science, Google Scholar
-
B. Roy , Multicriteria Methodology for Decision Aiding ( Kluwer Academic Publishers , 1996 ) . Crossref, Google Scholar -
B. Sarwar , Analysis of recommendation algorithms for e-commerce , Proc. ACM EC'00 ( 2000 ) , DOI: 10.1145/352871.352887 . Google Scholar -
V. Schickel-Zuber and B. Faltings , Hetereogeneous attribute utility model: a new approach for modelling user profiles for recommendation systems , Proc. Workshop Knowl. Discov. Web ( 2005 ) . Google Scholar -
C. Schmitt , D. Dengler and M. Bauer , The MAUT-machine: an adaptive recommender system , Proc. ABIS Workshop Adaptivität und Benutzermodellierung in interaktiven Softwaresystemen ( 2002 ) . Google Scholar -
M. Stolze and M. Stroebel , Dealing with learning in eCommerce product navigation and decision support: the teaching salesman problem , Proc. 2nd World Congr. Mass Custom. Person. ( 2003 ) . Google Scholar - Dec. Supp. Syst. 34, 127 (2002), DOI: 10.1016/S0167-9236(02)00076-3. Crossref, Web of Science, Google Scholar
-
P. Vincke , Multicriteria Decision-Aid ( J. Wiley , New York , 1992 ) . Google Scholar - J. Retailing 79, 183 (2003), DOI: 10.1016/S0022-4359(03)00034-4. Crossref, Web of Science, Google Scholar
-
K. Yu , Feature weighting and instance selection for collaborative filtering , Proc. 2nd Int. Workshop on Management of Information on the Web — Web Data and Text Mining (MIW'01) ( 2001 ) , DOI: 10.1109/DEXA.2001.953076 . Google Scholar - Int. J. Electr. Comm. 8(4), 115 (2004). Crossref, Web of Science, Google Scholar