Considering high utilities for time interval sequential pattern mining

The earlier ideas for time interval sequential pattern mining discovering the time interval pattern between two product items help managers promote the sale prediction of the items. For example, if most of the customers purchased product item A, and then they bought item B and C after r to s and t to u days respectively, the time interval between r to s and t to u days can be provided for managers to make marketing decision while predicting the purchasing time interval between A and B, as well as B and C based on the prediction. Nevertheless, the earlier research for the mining does not consider the significance of product items while mining the time-interval sequential patterns. This work implements the previous work and keeps time interval patterns with high utility in which the mining does not just consider the frequent item sets. This is intended to make the sequential pattern mining close to business behaviors. The experimental results show the differences between two mining approaches if considering item utility or item frequency for the purchased items.