February 2018
Intermediate to advanced
378 pages
10h 14m
English
The most famous algorithm for association rule learning is Apriori. It was proposed by Agrawal and Srikant in 1994. The input of the algorithm is a dataset of transactions where each transaction is a set of items. The output is a collection of association rules for which support and confidence are greater than some specified threshold. The name comes from the Latin phrase a priori (literally, "from what is before") because of one smart observation behind the algorithm: if the item set is infrequent, then we can be sure in advance that all its subsets are also infrequent.
You can implement Apriori with the following steps:
Read now
Unlock full access