作者: Guillem Lefait , Tahar Kechadi
DOI: 10.1109/ICDS.2010.47
关键词:
摘要: Knowledge on consumer habits is essential for companies to keep customers satisfied and provide them personalised services. We present a data mining architecture based clustering techniques help experts segment customer their purchase behaviours. In this architecture, diverse segmentation models are automatically generated evaluated with multiple quality measures. Some of these were selected given scores. Finally, the segments compared. This paper presents experimental results real-world set 10000 over 60 weeks 6 products. These experiments show that identified useful exploration discover interesting trends facilitated by use our architecture.