J Bobadilla et al. / Knowledge-Based Systems 23(2010)520-528 To complement the numerical values with additional non- 113] K. Goldberg. T. Perkins, Eigentaste: a constant time numerical information, focused on the arrangement of the votes collaborative filtering algorithm, Information Retrieval 4(2)(2001)133-151 of each pair of users, we manage to improve the results of Pearson [14] C.Hayes, P.Cunningham, Co ledge-Based Systems 17(2-4)(2004)131-138 correlation; for this purpose we use the Jaccard measure and the [151 J.L Herlocker, J.A. Konstan, A.l. Borchers, J.T. Riedl, An algorithmic framework traditional metric that best adapts to its characteristics: mean square differences. The new metric only operates with the data [161 ).L Herlocker, A Konstan, JT. Riedl, An empirical analysis of design choices in d-based collaborative filtering algorithms, Information Retrieval 5 (ratings)provided by the users of the recommender systems, and does not require any arbitrary parameters for adjustment or [171 J.L Herlocker, J.A. Konstan, J.T. Riedl, L.G. Terveen, Ev weighting. With the aim of achieving representative results the experi- [18)H. Ingoo, J.O. Kyong. H.R. Tae, The collaborative filtering recommendation ents have been carried out on three different recommender sys sed on SoM cluster-indexing CBR, Expert Systems with Applications 25 tems databases(MovieLens, FilmAffinity and NetFlix)which [19T Janner, C Schroth, Web 2.0 and SOA: converging concepts enabling the provide a sufficient volume and variety of data in order to offer reliable comparative results and general conclusions. The results [20] H Jinghua, w Kangning, F Shaohong. A survey of e-comme mender MovieLens and NetFlix(with range of votes [1.51), whilst the re- (211 M Knights, Web 2.0, IET Communications Engineer(2007)30-3 80214 ems confirm the integrity of the metric proposed when applied to ults do not improve that of Pearson correlation when it is applied [22] F. Kong. X Sun, s. Ye, A comparison of several algorithms for collaborative to FilmAffinity(with range of votes [ 1.10). filtering in startup stage, in: Proceedings of the IEEE Networking, Sensing and ontrol,2005,pp.25-28 Acknowledgements [24] G. Koutrica, B. Bercovitz, H Garci m2((204 Our acknowledgement to the GroupLens Research Group, and (25)P Li, S Yamada, A movie recommender system based on inductive learning. in ics and Intelligent Systems, 0.1109Ccs2004.1460433 Reference [26 K]. Lin, Building Web 2.0, Computer (2007)101-102. [271 G. Linden, B. Smith, J. York, Amazon. com recommendations, IEEE Internet systems: a survey of the state-c ossible extensions. IEEE in: 42nd Hawaii International Conference on System Sciences HICSS09, 2009. 如如m collaborative and content-based filtering. IEEE Intelligent Systems(2006)35- 30]Movielens,<http://www.movielens.org> [4]R Baraglia, F. Silvestri, An online recommender system for large web sites, L, An improvement to collaborative filtering for edings of the IEEE/IC/ACM International Conference on Web Modeling, Control and Automatisation, vol. 2005,pp.792-795,10.1109/McA2005.1631361 P. Pu, L Chen, Trust-inspiring explanation interfaces for recommender Heckerman, C Kadie, Empirical analysis of predicti (6)(2007)542-556. nding using form Uncertainty in Artificial Intelligence, Morgan Kaufmann, 1998, pp 43-52. [7 L Candillier, F. Meyer, M. Boulle, Comparing state-of-the-art collaborative [35] J. L Sanchez, F. Serradilla, E. Martinez, J Bobadilla, obile devices. Lecture Notes on echnologies DEST,2008,pp.432-436,10.1109/DEST20084635147 er Science4611(2007)1130-1139 [36]JB. Schafer, D. F ki, J.L Herlocker, S. Sen, Collaborative filtering [9 F. Hernandez, E. Gaudioso, Evaluation of recommender syst stems with Applications (35)(2008 [37 P. Symeonidis, A. Nanopoulos, Y. Manolopoulos, Providing justifications in [10] D.R. Fesenmaier. U. Gretzel. C. Knoblock, C. Paris, F. Ricci, S. Stabb, et al, recommender systems, IEEE Transactions on Systems, Man and Cybernetics, Intelligent systems for tourism, Intelligent Systems 17(6)(20 [38] W. Yuan, D Guan, Y.K. Lee, S. Lee, S. Hur, Improved recommender [11]I Fuyuki, T K Quan, H Shinichi. In mall-worldness of trust network e-Based Systems clustering items based on stability of user si Knowledge ns, Fuzzy Sets and Li. Recommendation base commendation algorithms: approaches anchored on human factors, collaborative filtering. Knowledge-Based Systems 22(1)(2009)105-114 Interacting with Computers 18 (3)(2006)410-431To complement the numerical values with additional non￾numerical information, focused on the arrangement of the votes of each pair of users, we manage to improve the results of Pearson correlation; for this purpose we use the Jaccard measure and the traditional metric that best adapts to its characteristics: mean square differences. The new metric only operates with the data (ratings) provided by the users of the recommender systems, and does not require any arbitrary parameters for adjustment or weighting. With the aim of achieving representative results the experi￾ments have been carried out on three different recommender sys￾tems databases (MovieLens, FilmAffinity and NetFlix) which provide a sufficient volume and variety of data in order to offer reliable comparative results and general conclusions. The results confirm the integrity of the metric proposed when applied to MovieLens and NetFlix (with range of votes [1..5]), whilst the re￾sults do not improve that of Pearson correlation when it is applied to FilmAffinity (with range of votes [1..10]). Acknowledgements Our acknowledgement to the GroupLens Research Group, and to the FilmAffinity and NetFlix companies. References [1] E. Adomavicius, A. Tuzhilin, Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions, IEEE Transactions on Knowledge and Data Engineering 17 (6) (2005) 734–749. [2] AICU, <http://aicu.eui.upm.es/aicu/doku.php?id=english>, <www.filmaffinity. com/research>. [3] N. Antonopoulus, J. Salter, CinemaScreen recommender agent: combining collaborative and content-based filtering, IEEE Intelligent Systems (2006) 35– 41. [4] R. Baraglia, F. Silvestri, An online recommender system for large web sites, in: Proceedings of the IEEE/WIC/ACM International Conference on Web Intelligence, 2004, pp. 199–205, 10.1109/WI.2004.10158. [5] J. Bobadilla, F. Serradilla, A. Hernando, Collaborative filtering adapted to recommender systems of e-learning, Knowledge-Based Systems 22 (2009) 261–265, doi:10.1016/j.knosys.2009.01.008. [6] J.S. Breese, D. Heckerman, C. Kadie, Empirical analysis of predictive algorithms for collaborative filtering, in: Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence, Morgan Kaufmann, 1998, pp. 43–52. [7] L. Candillier, F. Meyer, M. Boullé, Comparing state-of-the-art collaborative filtering systems, in: LNAI 4571, 2007, pp. 548–562. [8] S.B. Cho, J.H. Hong, M.H. Park, Location-based recommendation system using bayesian user’s preference model in mobile devices, Lecture Notes on Computer Science 4611 (2007) 1130–1139. [9] F. Hernández, E. Gaudioso, Evaluation of recommender systems: a new approach, Expert Systems with Applications (35) (2008) 790–804, doi:10.1016/j.eswa.2007.07.047. [10] D.R. Fesenmaier, U. Gretzel, C. Knoblock, C. Paris, F. Ricci, S. Stabb, et al., Intelligent systems for tourism, Intelligent Systems 17 (6) (2002) 53–66. [11] I. Fuyuki, T.K. Quan, H. Shinichi, Improving accuracy of recommender systems by clustering items based on stability of user similarity, in: Proceedings of the IEEE International Conference on Intelligent Agents, Web Technologies and Internet Commerce, 2006, pp. 61–61, 10.1109/CIMCA.2006.123. [12] G.M. Giaglis, G. Lekakos, Improving the prediction accuracy of recommendation algorithms: approaches anchored on human factors, Interacting with Computers 18 (3) (2006) 410–431. [13] K. Goldberg, T. Roeder, D. Gupta, C. Perkins, Eigentaste: a constant time collaborative filtering algorithm, Information Retrieval 4 (2) (2001) 133–151. [14] C. Hayes, P. Cunningham, Context boosting collaborative recommendations, Knowledge-Based Systems 17 (2–4) (2004) 131–138. [15] J.L. Herlocker, J.A. Konstan, A.l. Borchers, J.T. Riedl, An algorithmic framework for performing collaborative filtering, in: SIGIR, 1999, pp. 230–237. [16] J.L. Herlocker, J.A. Konstan, J.T. Riedl, An empirical analysis of design choices in neighborhood-based collaborative filtering algorithms, Information Retrieval 5 (2002) 287–310. [17] J.L. Herlocker, J.A. Konstan, J.T. Riedl, L.G. Terveen, Evaluating collaborative filtering recommender systems, ACM Transactions on Information Systems 22 (1) (2004) 5–53. [18] H. Ingoo, J.O. Kyong, H.R. Tae, The collaborative filtering recommendation based on SOM cluster-indexing CBR, Expert Systems with Applications 25 (2003) 413–423. [19] T. Janner, C. Schroth, Web 2.0 and SOA: converging concepts enabling the internet of services, in: IT Professional, 2007, pp. 36–41. [20] H. Jinghua, W. Kangning, F. Shaohong, A survey of e-commerce recommender systems, in: Proceedings of the International Conference on Service Systems and Service Management, 2007, pp. 1–5, 10.1109/ICSSSM.2007.4280214. [21] M. Knights, Web 2.0, IET Communications Engineer (2007) 30–35. [22] F. Kong, X. Sun, S. Ye, A comparison of several algorithms for collaborative filtering in startup stage, in: Proceedings of the IEEE Networking, Sensing and Control, 2005, pp. 25–28. [23] J.A. Konstan, B.N. Miller, J. Riedl, PocketLens: toward a personal recommender system, ACM Transactions on Information Systems 22 (3) (2004) 437–476. [24] G. Koutrica, B. Bercovitz, H. Garcia, FlexRecs: expresing and combining flexible recommendations, in: SIGMOD, 2009, pp. 745–757. [25] P. Li, S. Yamada, A movie recommender system based on inductive learning, in: Proceedings of the IEEE Conference on Cybernetics and Intelligent Systems, vol. 1, 2004, pp. 318–323, 10.1109/ICCIS.2004.1460433. [26] K.J. Lin, Building Web 2.0, Computer (2007) 101–102. [27] G. Linden, B. Smith, J. York, Amazon.com recommendations, IEEE Internet Computing (January–February) (2003) 76–80. [28] F. Loll, N. Pinkwart, Using collaborative filtering algorithms as eLearning tools, in: 42nd Hawaii International Conference on System Sciences HICSS‘09, 2009, pp. 1–10. [29] Y. Manolopoulus, A. Nanopoulus, A.N. Papadopoulus, P. Symeonidis, Collaborative recommender systems: combining effectiveness and efficiency, Expert Systems with Applications (34) (2007) 2995–3013, doi:10.1016/ j.eswa.2007.05.013. [30] MovieLens, <http://www.movielens.org>. [31] R. Nayak, L.T. Weng, Y. Xu, An improvement to collaborative filtering for recommender systems, in: Proceedings of the IEEE International Conference on Computational Intelligence for Modeling, Control and Automatisation, vol. 1, 2005, pp. 792–795, 10.1109/CIMCA.2005.1631361. [32] NetFlix, <www.netflixprize.com>. [33] P. Pu, L. Chen, Trust-inspiring explanation interfaces for recommender systems, Knowledge-Based Systems 20 (6) (2007) 542–556. [34] P.B. Ryan, D. Bridge, Collaborative recommending using formal concept analysis, Knowledge-Based Systems 19 (5) (2006) 309–315. [35] J.L. Sanchez, F. Serradilla, E. Martinez, J. Bobadilla, Choice of metrics used in collaborative filtering and their impact on recommender systems, in: Proceedings of the IEEE International Conference on Digital Ecosystems and Technologies DEST, 2008, pp. 432–436, 10.1109/DEST.2008.4635147. [36] J.B. Schafer, D. Frankowski, J.L. Herlocker, S. Sen, Collaborative filtering recommender systems, LNCS 4321, 2007, pp. 291–324. [37] P. Symeonidis, A. Nanopoulos, Y. Manolopoulos, Providing justifications in recommender systems, IEEE Transactions on Systems, Man and Cybernetics, Part A 38 (6) (2008) 1262–1272. [38] W. Yuan, D. Guan, Y.K. Lee, S. Lee, S.J. Hur, Improved trust-aware recommender system using small-worldness of trust networks, Knowledge-Based Systems 23 (3) (2010) 232–238. [39] R.R. Yager, Fuzzy logic methods in recommender systems, Fuzzy Sets and Systems 136 (2) (2003) 133–149. [40] J.M. Yang, K.F. Li, Recommendation based on rational inferences in collaborative filtering, Knowledge-Based Systems 22 (1) (2009) 105–114. 528 J. Bobadilla et al. / Knowledge-Based Systems 23 (2010) 520–528