Segmentation of the on-line shopping market using neural networks

摘要: Abstract The characterization and analysis of on-line customers' needs expectations, regarding the Internet as a new marketing channel, is considered prerequisite to realization expected growth consumer-oriented electronic commerce market. aim present study twofold: carry out an exploratory segmentation this market that can throw some light upon its structure, characterize shopping adoption process. Self-Organizing Map (SOM), unsupervised neural network model devised by Kohonen (Kohonen, T. (1982). Self-organized formation topologically correct feature maps. Biological Cybernetics, 43(1), 59–69; Kohonen, T., (1995). Self-organizing Berlin: Springer) will be used part tandem approach segmentation, which involves factor observable variables in data analyzed, prior clustering. SOM shown powerful visualization tool, able assist analysis, providing supervised methods with useful explanatory capabilities. It also applied, completely mode, discover clusters or segments naturally occur data. proposed flexible clustering accommodate both Finer Segmentation Normative approaches. Within latter, cluster-partition analysed, high-level customer profiles, potential interest marketers, are derived described terms.