Abstract
Along with the information increase on the Internet, there is a pressing need for online and real-time recommendation in commercial applications. This kind of recommendation attains results by combining both users’ historical data and their current behaviors. Traditional recommendation algorithms have high computational complexity and thus their reactions are usually delayed when dealing with large historical data. In this paper, we investigate the essential need of online and real-time processing in modern applications. In particular, to provide users with better online experience, this paper proposes an incremental recommendation algorithm to reduce the computational complexity and reaction time. The proposed algorithm can be considered as an online version of nonnegative matrix factorization. This paper uses matrix sketching and k-means clustering to deal with cold-start users and existing users respectively and experiments show that the proposed algorithm can outperform its competitors.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Song, Q., Cheng, J., Lu, H.: Incremental matrix factorization via feature space re-learning for recommendation system. In: RecSys, pp. 277–280 (2015)
Leng, C., Wu, J., Cheng, J., Bai, X., Lu, H.: Online sketching hashing. In: CVPR, pp. 2503–2511 (2015)
Liberty, E.: Simple and deterministic matrix sketching (2012). CoRR abs/1206.0594
Takacs, G., Pilaszy, I., Nemeth, B., Tikk, D.: Scalable collaborative filtering approaches for large recommender systems. J. Mach. Learn. Res. (JMLR) 10, 623–656 (2009)
Lommatzsch, A., Albayrak, S.: Real-time recommendations for user-item streams. In: SAC, pp. 1039–1046 (2015)
Brand, M.: Fast online SVD revision for lightweight recommender systems. In: SDM, pp. 37–46. SIAM (2003)
Gogna, A., Majumdar, A.: SVD free matrix completion with online bias correction for recommender systems. In: ICAPR, pp. 1–5 (2015)
Achakulvisut, T., Acuna, D.E., Ruangrong, T., Kording, K.P.: Science concierge: a fast content-based recommendation system for scientific publications (2016). CoRR abs/1604.01070
Burke, R.: Hybrid web recommender systems. In: Brusilovsky, P., Kobsa, A., Nejdl, W. (eds.) Adaptive Web 2007. LNCS, vol. 4321, pp. 377–408. Springer, Heidelberg (2007)
Rendle, S., Schmidt-Thieme, L.: Online-updating regularized kernel matrix factorization models for large-scale recommender systems. In: RecSys, pp. 251–258 (2008)
Bilge, A., Polat, H.: A scalable privacy-preserving recommendation scheme via bisecting k-means clustering. Inf. Process. Manage. 49(4), 912–927 (2013)
Subercaze, J., Gravier, C., Laforest, F.: Real-time, scalable, content-based Twitter users recommendation. Web Intell. 14(1), 17–29 (2016)
Lee, D.D., Seung, H.S.: Learning the parts of objects by nonnegative matrix factorization. Nature 401(6755), 788–791 (1999)
Acknowledgements
The work was supported by NSFC (U1435214, 61432008, 61503178, 61175042, 61403208, 61321491), Jiangsu Nature Science Foundation (JSNSF) (BK20150587, DE2015213).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing AG
About this paper
Cite this paper
Zhang, C., Wang, H., Yang, S., Gao, Y. (2016). Incremental Nonnegative Matrix Factorization Based on Matrix Sketching and k-means Clustering. In: Yin, H., et al. Intelligent Data Engineering and Automated Learning – IDEAL 2016. IDEAL 2016. Lecture Notes in Computer Science(), vol 9937. Springer, Cham. https://doi.org/10.1007/978-3-319-46257-8_46
Download citation
DOI: https://doi.org/10.1007/978-3-319-46257-8_46
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-46256-1
Online ISBN: 978-3-319-46257-8
eBook Packages: Computer ScienceComputer Science (R0)