Efficient Support Coupled Frequent Pattern Mining Over Progressive Databases

TL;DR

This paper introduces an efficient method for mining frequent sequential patterns with support in progressive databases, addressing real-world challenges like data stream dynamics and item support in market analysis.

Contribution

It proposes a novel progressive mining tree approach that effectively captures support-coupled sequential patterns in evolving datasets.

Findings

Efficiently mines frequent sequential patterns in dynamic databases.

Addresses real-world issues like data stream updates and item support.

Improves upon existing methods by integrating support considerations.

Abstract

There have been many recent studies on sequential pattern mining. The sequential pattern mining on progressive databases is relatively very new, in which we progressively discover the sequential patterns in period of interest. Period of interest is a sliding window continuously advancing as the time goes by. As the focus of sliding window changes, the new items are added to the dataset of interest and obsolete items are removed from it and become up to date. In general, the existing proposals do not fully explore the real world scenario, such as items associated with support in data stream applications such as market basket analysis. Thus mining important knowledge from supported frequent items becomes a non trivial research issue. Our proposed novel approach efficiently mines frequent sequential pattern coupled with support using progressive mining tree.