A Framework for Incremental Mining of Interesting Temporal Association Rules

Abstract

Listen

Translate article

In this paper, an incremental association rules mining algorithm is proposed that integrates interestingness criterion during the process of building the model called SUMA. One of the main features of the proposed framework is to capture the user background knowledge, which is monotonically augmented. The incremental model that reflects the changing data over the time and the user beliefs is attractive in order to make the over all KDD process more effective and efficient. The proposed framework is implemented and experiment it with some public datasets and found the results quite promising.