Abstract
Context-aware recommender systems improve context-free recommenders by exploiting the knowledge of the contextual situation under which a user experienced and rated an item. They use data sets of contextually-tagged ratings to predict how the target user would evaluate (rate) an item in a given contextual situation, with the ultimate goal to recommend the items with the best estimated ratings. This paper describes and evaluates a pre-filtering approach to context-aware recommendation, called distributional-semantics pre-filtering (DSPF), which exploits in a novel way the distributional semantics of contextual conditions to build more precise context-aware rating prediction models. In DSPF, given a target contextual situation (of a target user), a matrix-factorization predictive model is built by using the ratings tagged with the contextual situations most similar to the target one. Then, this model is used to compute rating predictions and identify recommendations for that specific target contextual situation. In the proposed approach, the definition of the similarity of contextual situations is based on the distributional semantics of their composing conditions: situations are similar if they influence the user’s ratings in a similar way. This notion of similarity has the advantage of being directly derived from the rating data; hence it does not require a context taxonomy. We analyze the effectiveness of DSPF varying the specific method used to compute the situation-to-situation similarity. We also show how DSPF can be further improved by using clustering techniques. Finally, we evaluate DSPF on several contextually-tagged data sets and demonstrate that it outperforms state-of-the-art context-aware approaches.
Similar content being viewed by others
Notes
South Tyrol Suggest is a mobile application currently available on the Google Play Store. See [https://play.google.com/store/apps/details?id=it.unibz.sts.android] (accessed June \(4{\mathrm{th}}\), 2014).
See [http://www.movielens.org] (accessed October \(27{\mathrm{th}}\), 2013).
See [http://www.librarything.com] (accessed October \(27{\mathrm{th}}\), 2013)
Note that this does not depend on the fact that the best semantic vector representation in each data set was selected (as described in the previous section) by using the direct measure of the similarity. In fact, the best semantic vector representation, i.e., either the user-based or the item-based (as mentioned previously) does not change if the similarity measure is changed.
In Tourism we used a slightly different variant that employs Singular Value Decomposition to reduce the dimensionality of the original item-based semantic vectors because, in this particular case, it improved significantly the results. (See Codina et al. (2013b) for more details about this variant).
References
Adomavicius, G., Mobasher, B., Ricci, F., Tuzhilin, A.: Context-aware recommender systems. AI Mag. 32(3), 67–80 (2011)
Adomavicius, G., Sankaranarayanan, R., Sen, S., Tuzhilin, A.: Incorporating contextual information in recommender systems using a multidimensional approach. ACM Trans. Inf. Syst. 23(1), 103–145 (2005)
Adomavicius, G., Tuzhilin, A.: Context-aware recommender systems. In: Recommender Systems Handbook, pp. 217–256 (2011)
Baltrunas, L., Amatriain, X.: Towards time-dependant recommendation based on implicit feedback. In: Proceedings of the 1st Workshop on Context-Aware Recommender Systems (CARS’09). October 22–25, 2009, New York City, USA (2009)
Baltrunas, L., Kaminskas, M., Ludwig, B., Moling, O., Ricci, F., Aydin, A., Lüke, K., Schwaiger, R.: InCarMusic: Context-aware music recommendations in a car. In: Proceedings of 12th International Conference (EC-Web’11), pp. 89–100. August 30 - September 1, 2011, Toulouse, France (2011)
Baltrunas, L., Ludwig, B., Peer, S., Ricci, F.: Context relevance assessment and exploitation in mobile recommender systems. Pers. Ubiquitous Comput. 16(5), 507–526 (2012)
Baltrunas, L., Ludwig, B., Ricci, F.: Matrix Factorization Techniques for Context Aware. In: Proceedings of the 5th ACM conference on Recommender systems (RecSys’11), pp. 301–304. October 23–27, 2011, Chicago, USA (2011)
Baltrunas, L., Ricci, F.: Context-dependent items generation in collaborative filtering. In: Proceedings of the 3th ACM conference on Recommender system (RecSys’09), pp. 245–248. October 22–25, 2009, New York City, USA (2009)
Baltrunas, L., Ricci, F.: Experimental evaluation of context-dependent collaborative filtering using item splitting. User Model. User-Adapt. Interact. 24(1–2), 7–34 (2014). doi:10.1007/s11257-012-9137-9
Bazire, M., Brezillon, P.: Understanding Context Before Using It. In: Proceedings of the 5th International Conference on Modeling and Using Context (CONTEXT’05), LNCS vol. 3554, pp.113–192. July 5-8, 2005, Paris, France. (2005)
Campos, P., Díez, F., Cantador, I.: Time-aware recommender systems: a comprehensive survey and analysis of existing evaluation protocols. User Model User-Adapt. Interact. 24(1–2), 67–119 (2014)
Codina, V., Ricci, F., Ceccaroni, L.: Exploiting the Semantic Similarity of Contextual Situations for Pre-filtering Recommendation. In: S. Carberry, S. Weibelzahl, A. Micarelli, & G. Semeraro (Eds.), Proceedings of the 21th International Conference on User Modeling, Adaptation, and Personalization (UMAP’13), pp. 165–177. June 10–14, Rome, Italy: Springer, Berlin Heidelberg (2013a)
Codina, V., Ricci, F., Ceccaroni, L.: Local Context Modeling with Semantic Pre-filtering. In: Proceedings of the 7th ACM conference on Recommender systems (RecSys’13), pp. 363–366. October 14–16, 2013, Hong Kong: ACM New York, NY, USA (2013b)
Cremonesi, P., Koren, Y., Turrin, R.: Performance of recommender algorithms on top-n recommendation tasks. In: Proceedings of the 4th ACM conference on Recommender systems (RecSys’10) pp. 39–46. September 23–26, Barcelona, Spain (2010)
Dourish, P.: What we talk about when we talk about context. Pers. Ubiquitous Comput. 8(1), 19–30 (2004)
Hayes, C., Cunningham, P.: Context boosting collaborative recommendations. Knowl. Based Syst. 17(2–4), 131–138 (2004)
Hidasi, B., Tikk, D.: Fast ALS-Based Tensor Factorization for Context-Aware Recommendation. In: Proceedings of the 2012 European conference on Machine Learning and Knowledge Discovery in Databases (KDD’12), pp. 67–82. August 12–16, 2012 Beijing, China (2012)
Karatzoglou, A., Amatriain, X., Baltrunas, L., Oliver, N.: Multiverse recommendation: N-dimensional tensor factorization for context-aware collaborative filtering. In: Proceedings of the 4th ACM conference on Recommender systems (RecSys’10), pp. 79–86. September 23–26, Barcelona, Spain (2010)
Koenigstein, N., Dror, G., Koren, Y.: Yahoo! music recommendations: modeling music ratings with temporal dynamics and item taxonomy. In: Proceedings of the 5th ACM conference on Recommender systems (RecSys’11), pp. 165–172. October 23–27, 2011, Chicago, USA (2011)
Koren, Y.: Collaborative filtering with temporal dynamics. Commun. ACM 53(4), 89 (2010). doi:10.1145/1721654.1721677
Koren, Y., Bell, R.: Advances in collaborative filtering. In: Recommender Systems Handbook, pp. 145–186 (2011)
Kurucz, M., Benczúr, A., Csalogány, K.: Methods for large scale SVD with missing values. In: Proceedings of KDD Cup and Workshop (held during KDD-2007), pp. 31–38. San Jose, California, USA (2007)
Molino, P.: Semantic models for answer re-ranking in question answering. In: Proceedings of the 36th international ACM SIGIR conference on research and development in information retrieval (SIGIR’13), Dublin, Ireland (2013)
Musto, C., Semeraro, G., Lops, P., de Gemmis, M.: Combining distributional semantics and entity linking for context-aware content-based recommendation. In: User modeling, adaptation, and personalization (UMAP’14), pp. 381–392. Aalborg, Denmark (2014)
Nelder, J., Mead, R.: A simplex method for function minimization. Comput. J. 7(4), 308–313 (1965)
Odić, A., Tkalčič, M., Tasic, J., Košir, A.: Predicting and detecting the relevant contextual information in a movie-recommender system. Interacting with Computers, 1–17 (2013)
Panniello, U., Tuzhilin, A., Gorgoglione, M.: Comparing context-aware recommender systems in terms of accuracy and diversity: which contextual modeling, pre-filtering and post-filtering methods perform the. User Model User-Adapt. Interact. 24(1–2), 35–65 (2014)
Panniello, U., Tuzhilin, A., Gorgoglione, M., Palmisano, C., Pedone, A.: Experimental comparison of pre-vs. post-filtering approaches in context-aware recommender systems. In: Proceedings of the 3th ACM conference on Recommender systems (RecSys’09), pp. 265–268. October 22–25, 2009, New York City, USA (2009)
Rajaraman, A., Ullman, J.: Clustering. In: Mining of massive datasets, pp. 239–278 (2012)
Rendle, S., Gantner, Z., Freudenthaler, C., Schmidt-Thieme, L.: Fast context-aware recommendations with factorization machines. In: Proceedings of the 34th international ACM SIGIR conference on Research and development in Information Retrieval (SIGIR ’11), pp. 635–644. July 24–27, New York, USA: ACM Press (2011)
Rubenstein, H., Goodenough, J.B.: Contextual correlates of synonymy. Commun. ACM 8(10), 627–633 (1965)
Shani, G., Gunawardana, A.: Evaluating Recommendation Systems. In: Recommender Systems Handbook, pp. 257–297 (2011)
Shi, Y., Karatzoglou, A., Baltrunas, L., Larson, M.: TFMAP : optimizing MAP for top-N context-aware recommendation. In: Proceedings of the 35th international ACM SIGIR conference on Research and development in Information Retrieval (SIGIR ’12), pp. 155–164. August 12–16, Portland, USA (2012)
Turney, P.D., Pantel, P.: From frequency to meaning: vector space models of semantics. J. Artif. Int. Res. 37(1), 141–188 (2010)
Zheng, Y., Burke, R., Mobasher, B.: Optimal feature selection for context-aware recommendation using differential relaxation. In: RecSys’12 workshop on context-aware recommender systems (CARS’12). Dublin, Ireland (2012)
Zheng, Y., Burke, R., Mobasher, B.: Recommendation with differential context weighting. In: Proceedings of the 21th international conference on user modeling, adaptation, and personalization (UMAP’13), pp. 152–164. June 10–14, 2013, Rome, Italy (2013a)
Zheng, Y., Burke, R., Mobasher, B.: The role of emotions in context-aware recommendation. In: recSys’13 workshop on human decision making in recommender systems, pp. 21–28. October 14–16, 2013, Hong Kong: ACM New York, NY, USA (2013b)
Acknowledgments
The research described in this paper is partly supported by the SuperHub and the Citclops European projects (FP7-ICT-2011-7, FP7-ENV-308469), and the Universitat Politècnica de Catalunya – BarcelonaTech (UPC) under an FPI-UPC Grant. The opinions expressed in this paper are those of the authors and are not necessarily those of SuperHub or Citclops projects’ partners.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
The following Table 10 describes the abbreviations used to identify the considered prediction model variants.
Rights and permissions
About this article
Cite this article
Codina, V., Ricci, F. & Ceccaroni, L. Distributional semantic pre-filtering in context-aware recommender systems. User Model User-Adap Inter 26, 1–32 (2016). https://doi.org/10.1007/s11257-015-9158-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11257-015-9158-2