International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 45 - Number 10 |
Year of Publication: 2012 |
Authors: Chetna Chand, Amit Thakkar, Amit Ganatra |
10.5120/6815-9169 |
Chetna Chand, Amit Thakkar, Amit Ganatra . Target Oriented Sequential Pattern Mining using Recency and Monetary Constraints. International Journal of Computer Applications. 45, 10 ( May 2012), 12-18. DOI=10.5120/6815-9169
Many approaches in constraint based sequential pattern mining have been proposed and most of them focus only on the concept of frequency, which means, if a pattern is not frequent, it is removed from further consideration. Frequency is a good indicator of the importance of a pattern but in real life, however, the environment may change constantly and patterns discovered from database may also change over time. Therefore, the users' recent behavior is not necessarily the same as the past ones and a pattern that occurs frequently in the past may never happen again in the future. So in this paper we have considered recency constraint to overcome this problem. Also we have considered one more constraint, monetary constraint since for making effective marketing strategies it is important to know the value of customer on the basis of what they are purchasing periodically and how much they are spending. So this motivates to consider monetary value of customers for targeting profitable customers. Along with that we have included the concept of mining only target oriented sequential patterns which satisfy RFM constraints to find the happening order of a concerned itemsets only, for taking effective marketing decisions.