Data Mining enables users to analyse, classify and discover correlations among data. One of the crucial tasks of this process is Association Rule Learning.
What Is Association Rule Learning (ARL)
An important part of data mining is anomaly detection, which is a procedure of search for items or events that do not correspond to a familiar pattern. These familiar patterns are termed anomalies and interpret critical and actionable data in various application fields. Association Rule Learning searches for association among variables.
This concept can be best understood with the supermarket example. In a supermarket, they infer data about the customer purchasing pattern for various products. With the help of association rule, the supermarket can distinguish which products are often bought together and this information can be used for marketing objectives.
How Does It Work
Association Rules work on the basis of if/then statements. These statements help to reveal associations between independent data in a database, relational database or other information repositories. These rules are used to identify the relationships between the objects which are usually used together.
The two primary patterns that association rules uses are support and confidence. The method finds similarities and rules formed by decomposing data for often used if/then patterns. Association rules are normally used to satisfy a user-specified minimum support and a use- specified minimum resolution simultaneously. There are various algorithms that are used to implement association rule learning.
Apriori algorithm is a standard algorithm in data mining. It is used for mining familiar item sets and relevant association rules. It is developed to perform on a database including a lot of transactions, for example, things brought by shoppers in a store.
It is very essential for effective Market Basket Analysis and it helps in understanding which items customers will buy together. It has also been used in the field of healthcare for the discovery of adverse drug reactions. It generates association rules that shows what all combinations of medications and patient characteristics which help in effective drug delivering.
Eclat algorithm is applied to achieve itemset mining. Itemset mining lets us obtain periodic patterns in data. For example, if a consumer buys butter, he would also buy bread.
The primary purpose of this algorithm is to use set intersections to compute the support of a candidate itemset avoiding the generation of subsets that does not exist in the prefix tree.
Eclat algorithm performs a depth-first search to count the columns. The Eclat algorithm, therefore, performs faster than the Apriori algorithm.
F-P Growth Algorithm
The Frequent Pattern (FP)-Growth classification is used with databases and not with streams. Apriori algorithm needs n+1 scans if a database is utilised, where n is the length of the longest model. By using the FP-Growth method, the number of scans of the complete database can be decreased to two.
The main applications of Association Rules are in data analysis, classification, cross-marketing, clustering, catalogue design, and loss-leader analysis and many others.