A survey of itemset mining

Abstract

Itemset mining is an important subfield of data mining, which consists of discovering interesting and useful patterns in transaction databases. The traditional task of frequent itemset mining is to discover groups of items (itemsets) that appear frequently together in transactions made by customers. Although itemset mining was designed for market basket analysis, it can be viewed more generally as the task of discovering groups of attribute values frequently cooccurring in databases. Because of its numerous applications in domains such as bioinformatics, text mining, product recommendation, e‐learning, and web click stream analysis, itemset mining has become a popular research area. This study provides an up‐to‐date survey that can serve both as an introduction and as a guide to recent advances and opportunities in the field. The problem of frequent itemset mining and its applications are described. Moreover, main approaches and strategies to solve itemset mining problems are presented, as well as their characteristics are provided. Limitations of traditional frequent itemset mining approaches are also highlighted, and extensions of the task of itemset mining are presented such as high‐utility itemset mining, rare itemset mining, fuzzy itemset mining, and uncertain itemset mining. This study also discusses research opportunities and the relationship to other popular pattern mining problems, such as sequential pattern mining, episode mining, subgraph mining, and association rule mining. Main open‐source libraries of itemset mining implementations are also briefly presented. WIREs Data Mining Knowl Discov 2017, 7:e1207. doi: 10.1002/widm.1207 This article is categorized under: Algorithmic Development > Association Rules Technologies > Association Rules

Frequent itemset mining: A 25 years review

“…The aim of this article is to show the improvements addressed during the last 25 years, that is, since the FIM task was first described (Agrawal et al, ). While some reviews have been already proposed in literature (Chee, Jaafar, Aziz, Hasan, & Yeoh, ; Fournier‐Viger et al, ), they are mainly focused on sequential exhaustive search approaches and on describing the algorithms for nonexpert users. In this sense, it is our understanding that an analysis from an expert point of view that involves any existing methodology (exhaustive and nonexhaustive search) on any architecture (centralized and parallel) is necessary to comprehend which improvements have been proposed over time.…”

“…Since 1993, when the first FIM algorithm was released (Agrawal et al, 1993), a special attention has been given to the performance of novel algorithms in this field (Borgelt, 2012;Fournier-Viger, Lin, Vo, Truong et al, 2017). Nowadays, 25 years later, extremely large datasets can be analyzed in a few seconds and this is not only a matter of novel architectures and hardware progresses but also a consequence of the proposed algorithmic solutions.…”

A Survey on High Utility Itemsets Mining

“…3. It follows breadth-first search approach which is quite costly in terms of memory utilization [41].…”

A survey of incremental high‐utility itemset mining

“…In many real‐world applications, data mining techniques are used to extract interesting patterns from databases, to support crucial decision‐making. Two fundamental tasks for revealing interesting relationships between items in transactional databases are frequent itemset mining (FIM) and association rule mining (ARM) (Agrawal, Imielinski, & Swami, ; Chen, Han, & Yu, ; Fournier‐Viger et al, ). The most well‐known ARM algorithms are Apriori (Agrawal & Srikant, ) and FP‐Growth (Han, Pei, Yin, & Mao, ).…”

“…To the best of our knowledge, this is the first survey on the mining task of incremental high‐utility itemset mining. The methods discussed in this article are not only important for iHUIM (Ahmed et al, ; Fournier‐Viger et al, ; Lin et al, ), but can also serve as inspiration for other data mining tasks (Fournier‐Viger et al, ), including incremental data mining (Hong et al, ) and dynamic data mining (Lin et al, ). The major contributions of this paper are threefold. A taxonomy of the most common approaches for mining HUIs in static databases, including Apriori‐based, tree‐based, projection‐based, hybrid, and other approaches, is presented.…”