Optimized RBF for CBR-Recommendation System

Abstract:

Recommendation systems are widely used in E-commerce to help their customers find products to purchase, with which an important problem is to efficiently search the contents with their demands, and have been attracting attention from quite a few researchers and practitioners from different fields. This paper proposes the CBR-recommender (Case-Based Reasoning) which is a comprehensive expression of human sense, logics and creativity, and can automatically acquire the user’s preferences from the process of adaptation or revision to satisfy the personalized needs; and we deploy radial basis function network (RBF) to control the system scale caused by the large amounts of data with high dimensions, whose performance is also superior with respect to the total time for satisfying a query Our experiments indicate that our mechanism is efficient since it is bounded by the number of neighbors and scalable because no global knowledge is required to be maintained.

• Periodical: Applied Mechanics and Materials (Volume 214)
• Pages: 568-572
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.214.568
• Online since: November 2012
• Authors: Jian Yang Li, Xiao Ping Liu, Rui Li
• Keywords: Automatic Acquisition, CBR-Recommender Modeling, Optimized RBF Networks, Similarity Measurement, System Construction

References

[1] Henrik Stormer. AAAI Workshop (2007), pp.92-99

[2] Kim H, Chan P K: Implicit Indicator for Interesting Web Pages,: (2005), pp.270-277

[3] Khalid Al-Kofahi, Peter Jackson etc. AAAI Workshop (2007), pp.9-18

[4] Jianyang Li, Rui Li, Jinbin Zheng, Zhihong Zeng, in: ICEIT 2010, (2010), pp.162-166

DOI: 10.1109/iceit.2010.5607776

[5] NI Zhiwei, CAI Qingsheng, JIA Ruiyu: Computer Engineering, Vol. 28(7). (2002), pp.12-15

[6] Lee Y, al: Berlin, Springer, (2004), pp.425-430

[7] Peng Zhao in: The University of Science and Technology, (2006).

[8] Choi J, Lee G, Um Y: Berlin, Springer, (2007)