Desmistificando o uso de Matrix Factorization em domínios de recomendação
Resumo
Apesar da popularidade de métodos de Matrix Factorization (MF) em Sistemas de Recomendação (SsR), não é clara a distinção prática entre eles em domínios reais. Neste contexto, este trabalho organiza métodos existentes, propondo uma organização taxonômica para a área, e caracteriza seus desempenhos, correlacionando seus sucessos da aplicação com características de cada domínio. Visando o uso prático desta pesquisa, apresentamos um arcabouço para caracterizaçãao de métodos de MF aplicados a SsR, útil para se contrastar o desempenho de distintos métodos. Análises sobre coleções reais demonstraram que recomendações geradas via SVD tendem a focar em itens não populares, enquanto recomendações oriundas do PCA focam em itens populares.
Referências
Adomavicius, G. and Tuzhilin, A. (2005). Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions. Knowledge and Data Engineering, IEEE Transactions on, 17(6):734–749.
Bell, R. M. and Koren, Y. (2007). Lessons from the netflix prize challenge. ACM SIGKDD Explorations Newsletter, 9(2):75–79.
Funk, S. (2006). Netflix update: Try this at home (december 2006). UR L http://sifter.org/˜ simon/journal/20061211. html.
Goldberg, K., Roeder, T., Gupta, D., and Perkins, C. (2001). Eigentaste: A constant time collaborative filtering algorithm. Information Retrieval, 4(2):133–151.
Koren, Y. (2008). Factorization meets the neighborhood: a multifaceted collaborative filtering model. In Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 426–434. ACM.
Koren, Y. (2010). Factor in the neighbors: Scalable and accurate collaborative filtering. ACM Transactions on Knowledge Discovery from Data (TKDD), 4(1):1.
Koren, Y., Bell, R., and Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, (8):30–37.
Paterek, A. (2007). Improving regularized singular value decomposition for collaborative filtering. In Proceedings of KDD cup and workshop, volume 2007, pages 5–8.
Sarwar, B., Karypis, G., Konstan, J., and Riedl, J. (2000). Application of dimensionality reduction in recommender system-a case study. Technical report, DTIC Document.
Sharifi, Z., Rezghi, M., and Nasiri, M. (2013). New algorithm for recommender systems based on singular value decomposition method. In Computer and Knowledge Engineering (ICCKE), 2013 3th International eConference on, pages 86–91. IEEE.
Silva, N., Neves, A., Rocha, R. C., and Mourão, F. (2015). Sobre as diferenças de aplicação entre svd e pca: Um estudo pragmático. In Anais do XXXV Congresso da Sociedade Brasileira de Computação.
Yin, C.-X. and Peng, Q.-K. (2012). A careful assessment of recommendation algorithms related to dimension reduction techniques. Knowledge-Based Systems, 27:407–423.
Bell, R. M. and Koren, Y. (2007). Lessons from the netflix prize challenge. ACM SIGKDD Explorations Newsletter, 9(2):75–79.
Funk, S. (2006). Netflix update: Try this at home (december 2006). UR L http://sifter.org/˜ simon/journal/20061211. html.
Goldberg, K., Roeder, T., Gupta, D., and Perkins, C. (2001). Eigentaste: A constant time collaborative filtering algorithm. Information Retrieval, 4(2):133–151.
Koren, Y. (2008). Factorization meets the neighborhood: a multifaceted collaborative filtering model. In Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 426–434. ACM.
Koren, Y. (2010). Factor in the neighbors: Scalable and accurate collaborative filtering. ACM Transactions on Knowledge Discovery from Data (TKDD), 4(1):1.
Koren, Y., Bell, R., and Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, (8):30–37.
Paterek, A. (2007). Improving regularized singular value decomposition for collaborative filtering. In Proceedings of KDD cup and workshop, volume 2007, pages 5–8.
Sarwar, B., Karypis, G., Konstan, J., and Riedl, J. (2000). Application of dimensionality reduction in recommender system-a case study. Technical report, DTIC Document.
Sharifi, Z., Rezghi, M., and Nasiri, M. (2013). New algorithm for recommender systems based on singular value decomposition method. In Computer and Knowledge Engineering (ICCKE), 2013 3th International eConference on, pages 86–91. IEEE.
Silva, N., Neves, A., Rocha, R. C., and Mourão, F. (2015). Sobre as diferenças de aplicação entre svd e pca: Um estudo pragmático. In Anais do XXXV Congresso da Sociedade Brasileira de Computação.
Yin, C.-X. and Peng, Q.-K. (2012). A careful assessment of recommendation algorithms related to dimension reduction techniques. Knowledge-Based Systems, 27:407–423.
Publicado
04/07/2016
Como Citar
SILVA, Nícollas; ROCHA, Leonardo; MOURÃO, Fernando.
Desmistificando o uso de Matrix Factorization em domínios de recomendação. In: CONCURSO DE TRABALHOS DE INICIAÇÃO CIENTÍFICA DA SBC (CTIC-SBC), 35. , 2016, Porto Alegre.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2016
.
p. 71-80.
