The Data Mining Forum 
IMPORTANT: This is the old Data Mining forum.
I keep it online so that you can read the old messages.
Please post your new messages in the new forum: https://forum2.philippe-fournier-viger.com/index.php
Association Rules Mining didn't work
Posted by:
dqanhtriet
Date: March 16, 2021 05:10PM
Hi all,
Thanks for spmf software, I can do more with my research about Association Rule Hiding, and I've used spmf to mine closed association rules, but it didn't work, atleast with mushroom, retail, accident dataset I downloaded from datasets page. Can you show me the way to do it? It worked with chess dataset.
This is the result of dataset mushroom.
Algorithm is running... (05:54:49 PM)
============= FP-Close v0.96r14 - STATS =============
Transactions count from database : 8416
Max memory usage: 8.153923034667969 mb
Closed frequent itemset count : 0
Total time ~ 99 ms
===================================================
============= ASSOCIATION RULE GENERATION v2.19- STATS =============
Number of association rules generated : 0
Total time ~ 0 ms
===================================================
Thanks for reading.
Thanks for spmf software, I can do more with my research about Association Rule Hiding, and I've used spmf to mine closed association rules, but it didn't work, atleast with mushroom, retail, accident dataset I downloaded from datasets page. Can you show me the way to do it? It worked with chess dataset.
This is the result of dataset mushroom.
Algorithm is running... (05:54:49 PM)
============= FP-Close v0.96r14 - STATS =============
Transactions count from database : 8416
Max memory usage: 8.153923034667969 mb
Closed frequent itemset count : 0
Total time ~ 99 ms
===================================================
============= ASSOCIATION RULE GENERATION v2.19- STATS =============
Number of association rules generated : 0
Total time ~ 0 ms
===================================================
Thanks for reading.
Re: Association Rules Mining didn't work
Posted by:
webmasterphilfv
Date: March 17, 2021 06:07AM
Hi,
Have you tried lowering the parameters ? (minimum support and minimum confidence?)
The parameter like minimum support must be set to very different values for each datasets.
For example, on some datasets, there might be 0 itemsets for minsup =0.4 while for another dataset, there might be millions of itemsets for minsup =0.9999
The best is to start with a high value and gradually decrease until you find enough rules.
For the minconf parameter, also, dont set it to a very high value. You may try some small value and then increase if you have too many rules.
Have you tried lowering the parameters ? (minimum support and minimum confidence?)
The parameter like minimum support must be set to very different values for each datasets.
For example, on some datasets, there might be 0 itemsets for minsup =0.4 while for another dataset, there might be millions of itemsets for minsup =0.9999
The best is to start with a high value and gradually decrease until you find enough rules.
For the minconf parameter, also, dont set it to a very high value. You may try some small value and then increase if you have too many rules.