Association rules are "if-then" statements, that help to show the probability of relationships between data items, within large data sets in various types of databases. Association rule mining has a number of applications and is widely used to help discover sales correlations in transactional data or in medical data sets.
Use cases for association rules
In data science, association rules are used to find correlations and co-occurrences between data sets. They are ideally used to explain patterns in data from seemingly independent information repositories, such as relational databases and transactional databases. The act of using association rules is sometimes referred to as "association rule mining" or "mining associations."Content Continues Below
Below are a few real-world use cases for association rules:
- Medicine. Doctors can use association rules to help diagnose patients. There are many variables to consider when making a diagnosis, as many diseases share symptoms. By using association rules and machine learning-fueled data analysis, doctors can determine the conditional probability of a given illness by comparing symptom relationships in the data from past cases. As new diagnoses get made, the machine learning model can adapt the rules to reflect the updated data.
- Retail. Retailers can collect data about purchasing patterns, recording purchase data as item barcodes are scanned by point-of-sale systems. Machine learning models can look for co-occurrence in this data to determine which products are most likely to be purchased together. The retailer can then adjust marketing and sales strategy to take advantage of this information.
- User experience (UX) design. Developers can collect data on how consumers use a website they create. They can then use associations in the data to optimize the website user interface -- by analyzing where users tend to click and what maximizes the chance that they engage with a call to action, for example.
- Entertainment. Services like Netflix and Spotify can use association rules to fuel their content recommendation engines. Machine learning models analyze past user behavior data for frequent patterns, develop association rules and use those rules to recommend content that a user is likely to engage with, or organize content in a way that is likely to put the most interesting content for a given user first.
How association rules work
Association rule mining, at a basic level, involves the use of machine learning models to analyze data for patterns, or co-occurrences, in a database. It identifies frequent if-then associations, which themselves are the association rules.
An association rule has two parts: an antecedent (if) and a consequent (then). An antecedent is an item found within the data. A consequent is an item found in combination with the antecedent.
Association rules are created by searching data for frequent if-then patterns and using the criteria support and confidence to identify the most important relationships. Support is an indication of how frequently the items appear in the data. Confidence indicates the number of times the if-then statements are found true. A third metric, called lift, can be used to compare confidence with expected confidence, or how many times an if-then statement is expected to be found true.
Association rules are calculated from itemsets, which are made up of two or more items. If rules are built from analyzing all the possible itemsets, there could be so many rules that the rules hold little meaning. With that, association rules are typically created from rules well-represented in data.
Measures of the effectiveness of association rules
The strength of a given association rule is measured by two main parameters: support and confidence. Support refers to how often a given rule appears in the database being mined. Confidence refers to the amount of times a given rule turns out to be true in practice. A rule may show a strong correlation in a data set because it appears very often but may occur far less when applied. This would be a case of high support, but low confidence.
Conversely, a rule might not particularly stand out in a data set, but continued analysis shows that it occurs very frequently. This would be a case of high confidence and low support. Using these measures helps analysts separate causation from correlation, and allows them to properly value a given rule.
A third value parameter, known as the lift value, is the ratio of confidence to support. If the lift value is a negative value, then there is a negative correlation between datapoints. If the value is positive, there is a positive correlation, and if the ratio equals 1, then there is no correlation.
Association rule algorithms
Popular algorithms that use association rules include AIS, SETM, Apriori and variations of the latter.
With the AIS algorithm, itemsets are generated and counted as it scans the data. In transaction data, the AIS algorithm determines which large itemsets contained a transaction, and new candidate itemsets are created by extending the large itemsets with other items in the transaction data.
The SETM algorithm also generates candidate itemsets as it scans a database, but this algorithm accounts for the itemsets at the end of its scan. New candidate itemsets are generated the same way as with the AIS algorithm, but the transaction ID of the generating transaction is saved with the candidate itemset in a sequential data structure. At the end of the pass, the support count of candidate itemsets is created by aggregating the sequential structure. The downside of both the AIS and SETM algorithms is that each one can generate and count many small candidate itemsets, according to published materials from Dr. Saed Sayad, author of Real Time Data Mining.
With the Apriori algorithm, candidate itemsets are generated using only the large itemsets of the previous pass. The large itemset of the previous pass is joined with itself to generate all itemsets with a size that's larger by one. Each generated itemset with a subset that is not large is then deleted. The remaining itemsets are the candidates. The Apriori algorithm considers any subset of a frequent itemset to also be a frequent itemset. With this approach, the algorithm reduces the number of candidates being considered by only exploring the itemsets whose support count is greater than the minimum support count, according to Sayad.
Uses of association rules in data mining
In data mining, association rules are useful for analyzing and predicting customer behavior. They play an important part in customer analytics, market basket analysis, product clustering, catalog design and store layout.
Programmers use association rules to build programs capable of machine learning. Machine learning is a type of artificial intelligence (AI) that seeks to build programs with the ability to become more efficient without being explicitly programmed.
Examples of association rules in data mining
A classic example of association rule mining refers to a relationship between diapers and beers. The example, which seems to be fictional, claims that men who go to a store to buy diapers are also likely to buy beer. Data that would point to that might look like this:
A supermarket has 200,000 customer transactions. About 4,000 transactions, or about 2% of the total number of transactions, include the purchase of diapers. About 5,500 transactions (2.75%) include the purchase of beer. Of those, about 3,500 transactions, 1.75%, include both the purchase of diapers and beer. Based on the percentages, that large number should be much lower. However, the fact that about 87.5% of diaper purchases include the purchase of beer indicates a link between diapers and beer.
While the concepts behind association rules can be traced back earlier, association rule mining was defined in the 1990s, when computer scientists Rakesh Agrawal, Tomasz Imieliński and Arun Swami developed an algorithm-based way to find relationships between items using point-of-sale (POS) systems. Applying the algorithms to supermarkets, the scientists were able to discover links between different items purchased, called association rules, and ultimately use that information to predict the likelihood of different products being purchased together.
For retailers, association rule mining offered a way to better understand customer purchase behaviors. Because of its retail origins, association rule mining is often referred to as market basket analysis.
As advances in data science, AI and machine learning, have occurred since the original use case for association rules -- and more devices generate data -- association rules can be used in wider breadth of use cases. More data is being generated, meaning more applications for association rules. AI and machine learning allow for larger and more complex data sets to be analyzed and mined for association rules.