by Yanchang Zhao, RDataMining.com
The technique of association rules is widely used for retail basket analysis, as well as in other applications to find assocations between itemsets and between sets of attribute-value pairs. It can also be used for classification by using rules with class labels on the right-hand side. It is even used for outlier detection with rules indicating infrequent/abnormal association.
Below are some free online resources on association rule mining with R and also documents on the basic theory behind the technique.
1. My R example and document on association rule mining, redundancy removal and rule interpretation
– http://www.rdatamining.com/examples/association-rules
– Chapter 9 Association Rules in book “R and Data Mining: Examples and Studies”
http://www.rdatamining.com/docs/RDataMining.pdf
2. Vignettes for mining and visualizing association rules
– Introduction to arules: A computational environment for mining association rules and frequent item sets
http://cran.csiro.au/web/packages/arules/vignettes/arules.pdf
– Visualizing Association Rules: Introduction to arulesViz
http://cran.csiro.au/web/packages/arulesViz/vignettes/arulesViz.pdf
3. Some R implementations of association rule algorithms:
– “Data Mining Algorithms In R: Frequent Pattern Mining” at
http://en.wikibooks.org/wiki/Data_Mining_Algorithms_In_R/Frequent_Pattern_Mining
4. Lecture Notes (slides in PDF) on the theory of association rules
– http://www-users.cs.umn.edu/~kumar/dmbook/dmslides/chap6_basic_association_analysis.pdf
– http://www-users.cs.umn.edu/~kumar/dmbook/dmslides/chap7_extended_association_analysis.pdf
5. A book chapter on Association Rules
http://www-users.cs.umn.edu/~kumar/dmbook/ch6.pdf
6. A comparison of over 20 interestingness measures for association rules:
– “Selecting the right objective measure for association analysis”
http://www.cse.msu.edu/~ptan/papers/IS.pdf
7. A collection of links to online resources on data mining (and R), not limitted to association rules:
– http://www.rdatamining.com/resources/onlinedocs
Pingback: Kuka käyttää tiedonlouhintaa ja mitä työkalua? « Olipa kerran Bigdata
Hi,
Thank you for your examples. They are very helpful.
For Equations 9.1, 9.3 and 9.5 of http://www.rdatamining.com/docs/RDataMining.pdf, should P(A union B) be P(A and B)?
Wendy
Thanks, Wendy. A and B are two item sets, so a set union is used there. It is the same as p(A and b).
how to implement association rule mining in hypertext and discovery of frequent episodes in event sequence
using r or the rapid miner