Knowledge discovery of weighted RFM-QR sequential patterns with multi time interval from customer sequence database
Chandni Naik, Ankit Kharwar · 2015
Sequential pattern mining is valuable approach to uncover consumer buying behaviour from huge sequence database. Weather prediction, web log analysis, stock market analysis, scientific research, sales analysis, and so on are the application of sequential pattern mining. The pattern that is recent and profitable can't discover by conventional sequential pattern mining. So, RFM-based sequential pattern mining methodology is established. Purchasing patterns that are recently active and profitable is discovered by RFM-based sequential pattern mining; but this methodology does not discover the time interval between each and every item as well as quantity of items in the pattern. It also doesn't offer the patterns as per the user preference. To discover a time interval and quantity of items, and patterns according to user preference, we proposed a RFM-TI-Q-RANK algorithm. The empirical result reveals that the proposed technique can find out more precious patterns than RFM-based sequential pattern mining.