Using a Clustering Genetic Algorithm to Support Customer Segmentation for Personalized Recommender Systems
This study proposes novel clustering algorithm based on genetic algorithms (GAs) to carry out a segmentation of the online shopping market effectively. In general, GAs are believed to be effective on NP-complete global optimization problems and they can provide good sub-optimal solutions in reasonable time. Thus, we believe that a clustering technique with GA can provide a way of finding the relevant clusters. This paper applies GA-based K-means clustering to the real-world online shopping market segmentation case for personalized recommender systems. In this study, we compare the results of GA-based K-means to those of traditional K-means algorithm and self-organizing maps. The result shows that GA-based K-means clustering may improve segmentation performance in comparison to other typical clustering algorithms.
KeywordsCluster Algorithm Recommender System Initial Seed Market Segmentation Uniform Crossover
Unable to display preview. Download preview PDF.
- 1.Bradley, P.S., Fayyad, U.M.: Refining Initial Points for K-means Clustering. In: Proc. of the 15th International Conference on Machine Learning, pp. 91–99 (1998)Google Scholar
- 2.Dibb, S., Simkin, L.: The Market Segmentation Workbook: Target Marketing for Marketing Managers, Routledge, London (1995)Google Scholar
- 4.Hertz, J., Krogh, A., Palmer, R.G.: Introduction to the Theory Neural Computation. Addison-Wesley, Reading (1991)Google Scholar
- 5.Kehoe, C., Pitkow, J., Rogers, J.D.: Ninth GVU.s WWW User Survey (1998), http://www.gvu.gatech.edu/user_surveys/survey-1998-04/
- 6.Kim, K.: Toward Global Optimization of Case-Based Reasoning Systems for Financial Forecasting. Applied Intelligence (2004) (forthcoming)Google Scholar
- 12.Wong, F., Tan, C.: Hybrid Neural, Genetic and Fuzzy Systems. In: Deboeck, G.J. (ed.) Trading On The Edge, pp. 243–261. Wiley, New York (1994)Google Scholar