Mining association rules between sets of items in large databases

By Rakesh Agrawal, Tomasz Imielinski, Arun N. Swami. 1993

We are given a large database of customer transactions. Each transaction consists of items purchased by a customer in a visit. We present an efficient algorithm that generates all significant association rules between items in the database. The algorithm incorporates buffer management and novel estimation and pruning techniques. We also present results of applying this algorithm to sales data obtained from a large retailing company, which shows the effectiveness of the algorithm.

Subfields: Information Retrieval & Search, Algorithms & Complexity

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Retrieval & Search

In Algorithms & Complexity

Citations in the corpus

Black-Box vs. White-Box: Understanding Their Advantages and Weaknesses From a Practical Point of View (2019, background): “Such rule can be used for decisions about marketing activities [108].”
Explainable Recommendation: A Survey and New Perspectives (2018, methodology): “The most frequently used data mining technique for explainable recommendation is association rule mining (Agrawal et al. [1993], Agarwal et al. [1994]), and a very classical example is the “beer-diaper” recommendation originated from data mining research. For example, Mobasher et al. [2001] leverage”
Interpretable Decision Sets: A Joint Framework for Description and Prediction (2016, methodology): “In practice, we follow a two step approach where we first use frequent itemset mining [1] to extract itemsets S and then apply our learning method to select a set of rules that form an interpretable decision set.”
Towards Integrative Machine Learning and Knowledge Extraction (2017, methodology): “address parser [20], Google Maps geocoding APIs [21], Geocoder.”
World-scale mining of objects and events from community photo collections (2008, methodology): “More precisely, as shown in [1], the problem can be formulated as follows.”
Pruning and summarizing the discovered associations (1999, background): “With a target attribute, we can express an itemset as follows (instead of a set of items as in [2]): X → y where y is an item (or a value) of the target attribute, and X is a set of items from the rest of the attributes.”
Machine Learning and Data Mining Methods in Diabetes Research (2017, background): “The most well-known association rule discovery algorithm is Apriori , proposed in 1994 by Rakesh Agrawal [8].”
Data Mining in Healthcare and Biomedicine: A Survey of the Literature (2012, background): “From the time Agrawl and his colleagues at IBM Almaden Research Center introduced a novel association rule algorithm called Apriori [ 75 , 76], association mining has received significant attention.”
Scalable Algorithms for Association Mining (2000, unknown): “A lattice L is said to be distributive if for all X; Y ; Z 2 L , Z (cid:134) .”
Data Mining for Internet of Things: A Survey (2014, background): “For this reason, the classification of mining technologies, possible combinations, and simple examples will be given in the following sections to show one of the many possible ways to combine mining technologies into a single system.”
The Convolutional Tsetlin Machine (2019, background): “One prominent example is frequent itemset mining for association rule learning [2], for instance applied to predicting sequential events [35, 24].”
Association Rules Mining: A Recent Overview (2006, methodology): “Association rule mining, one of the most important and well researched techniques of data mining, was first introduced in [1].”
Privacy preserving mining of association rules (2002, background): “• categorical data instead of numerical data, and • association rule mining [ 4 ] instead of classification.”
Detecting Novel Network Intrusions Using Bayes Estimators (2001, background): “The task of mining association rules, first presented in [2] consists in deriving a set of rules in the form of X −→ Y where X and Y are sets of attribute-values, with X ⋂ Y = ∅ and ‖Y ‖ = 1.”