I want a practical example on frequent itemset mining by using the Apriori algorithm, which can show me the the transaction database and minimum support threshold, and moreover, the numbers of candidate k-itemset and frquent k-itemsets for all k. Are there some papers which have given some information about this?
A step-by-step example can be found here:
http://www3.cs.stonybrook.edu/~cse634/lecture_notes/07apriori.pdf
Some papers:
http://www.robotics.stanford.edu/~ronnyk/realWorldAssocPoster.pdf
http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.25.3707
https://www.researchgate.net/publication/287316286_A_human-centered_approach_for_mining_hybrid-dimensional_association_rules
Real world datasets:
https://archive.ics.uci.edu/ml/datasets/Online+Retail
https://www.researchgate.net/publication/303840377_ces_hybrid_real_world_dataset_for_association_rule_mining?ev=prf_pub
Article A human-centered approach for mining hybrid-dimensional asso...
Data ces_hybrid (real world dataset for association rule mining)
Hi, Eduardo Corrêa Gonçalves.
The dataset and papers you provided really help me a lot.
However, are there some more direct papers which could list the the number of candidate itemsets and the number frequent itemsets of each iteration when running frequent itemsets mining algorithm for a real-world transaction database? For example, is there a paper proving the information like this: in a transaction database with 10000 transactions and 1000 items, there are 1000 candidate 1-itemsets, 100 frequent 1-itemsets, 4950 candidate 2-itemsets, 200 frequent 2-itemsets,...., 30 candidate 8-itemsets and 2 frequent 8-itemsets (stop) ?
Thank you very much.
Figures 2 and 3 of the below paper present the number of frequent itemsets x minimum support for 2 datasets (not too much, but it may be useful):
http://www.robotics.stanford.edu/~ronnyk/realWorldAssocPoster.pdf
- Badges
- Science topic
- Similar topics
- Computer Science
- Data Mining