Semantic Tags and Neutral Network Based Personalized Advertisement Recommendation System for Movies

Abstract:

With the rapid development of information especially internet technology, people have to choose the most suitable goods without any experience, so the recommendation system is seriously required. Yet no research on advertisement recommendation system for movie play is presented. Regarding this problem, the paper introduces the theory of semantic computing and annotates the semantic tags from the movie slices and the candidate advertisements, the potential preferences on them are predicted with neutral network model trained by some data set predefined. The user preference model and the predicting workflow are described in detail. Finally, the MovieLens dataset is employed to validate the validity of the system designed. The results of simulation experiments prove that the technology proposed can not only satisfy the requirement of matched advertisement recommendation but also outperform the traditional collaborative filtering algorithm.

• Periodical: Advanced Materials Research (Volumes 488-489)
• Pages: 1727-1731
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.488-489.1727
• Online since: March 2012
• Authors: Wei Du, Jun Liang Chen
• Keywords: Advertisement Recommendation, Recommendation System, Semantic Computing

References

[1] Huiyi, T., G. Junfei, and L. Yong. E-learning recommendation system[C]. 2008. Wuhan, Hubei, China: IEEE Computer Society.

[2] Baraglia, R., et al. A privacy preserving Web recommender system[C]. 2006. Dijon, France: Association for Computing Machinery.

[3] Chelcea, S., G. Gallais, and B. Trousse. A personalized recommender system for travel information Recommandations personnalisees pour la recherche d'information facilitant les deplacements[C]. 2004. Nice, France: Association for Computing Machinery.

DOI: 10.1145/1050873.1050905

[4] Agarwal, N., et al. Research paper recommander systems: A subspace clustering approach[C]. 2005. Hangzhou, China: Springer Verlag.

[5] Adomavicius, G. and A. Tuzhilin, Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions[J]. IEEE Transactions on Knowledge and Data Engineering, 2005. 17(6): pp.734-749.

DOI: 10.1109/tkde.2005.99