Abstract
While interactive television enables a new user-centered TV mode, catering to the tastes of TV users is one of the most critical tasks in delivering interactive TV experience. It faces two key challenges. First, the user behaviors on TV are much sparser than those of the internet users, thus making the modeling of user preferences more challenging. Second, an TV account is usually associated with multiple individuals in a family, making it difficult to discriminate the preferences of individual family members. In this paper, we thus propose a novel Clustering-Coupled Topic Model (CCTM), which characterizes user profile only by analyzing user viewing behaviors without any program metadata. This model clusters the users into different groups, then access the group preference for program recommendation by coupling the interest of different users in the same cluster group. We validate the performance of the CCTM with real-world data from a national interactive TV program. The experimental results have demonstrated that the CCTM can reasonably extract the users’ potential preference, which is further leveraged to recommend programs to the users.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Jan, D.V., Peters, O., Heuvelman, A.: Interactive television or enhanced televisiion? The Dutch users interest in applications of ITV via set-top boxes. In: Annual Meeting of the International Communication Association ICA (2017)
Cho, J.H., Sah, Y.J., Ryu, J.: A new content-related advertising model for interactive television. In: IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, pp. 1–9. IEEE (2008)
Branch, P., Egan, G., Tonkin, B.: Modeling interactive behavior of a video based multimedia system. In: Proceedings of the IEEE International Conference on Communications, pp. 978–982 (1999)
Gopalakrishnan, V., Jana, R., Knag, R., Ramakrishnan, K., Swayne, D., Vaishampayan, V.: Characterizing interactive behavior in a large-scale operational IPTV environment. In: Proceedings IEEE INFOCOM 2010, pp. 1–5 (2010)
Zhang, Y., Chen, W., Zha, H., et al.: A time-topic coupled LDA model for IPTV user behaviors. IEEE Trans. Broadcasting 61(1), 56–65 (2015)
Blei, D.M., Ng, A.Y., Jordan, M.I.: Latent dirichlet allocation. J. Mach. Learn. Res. 3, 9931022 (2003)
Aggarwal, C.C., Zhai, C.: A survey of text clustering algorithms. In: Aggarwal, C., Zhai, C. (eds.) Mining Text Data. Springer, Boston (2012). https://doi.org/10.1007/978-1-4614-3223-4_4
Ng, A.Y., Jordan, M.I., Weiss, Y., et al.: On spectral clustering: analysis and an algorithm. In: Advances in Neural Information Processing Systems, vol. 2, pp. 849–856 (2002)
Xie, P., Xing, E.P.: Integrating document clustering and topic modeling. In: Proceedings of the 29th International Conference on Uncertainty in Artificial Intelligence (2013)
Wainwright, M.J., Jordan, M.I.: Graphical models, exponential families, and variational inference. Found. Trends@ Mach. Learn. 1(1–2), 1–305 (2008)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Jiang, Z. et al. (2018). Cold-Start Group Profiling with a Clustering-Coupled Topic Model. In: Zhai, G., Zhou, J., Yang, X. (eds) Digital TV and Wireless Multimedia Communication. IFTC 2017. Communications in Computer and Information Science, vol 815. Springer, Singapore. https://doi.org/10.1007/978-981-10-8108-8_31
Download citation
DOI: https://doi.org/10.1007/978-981-10-8108-8_31
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-8107-1
Online ISBN: 978-981-10-8108-8
eBook Packages: Computer ScienceComputer Science (R0)