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
Result rules not match Apriori and FP-Growth
Posted by:
Anonymous_itop
Date: August 28, 2014 10:58AM
I,m run apriori and FP-Growth with data plants.data
http://archive.ics.uci.edu/ml/datasets/Plants
with min-conf = 0.92. But with apriori and FP-Growth have difference number of rules. With Apriori it have 212 Rules and FP-Growth have 211 Rules.
How is it difference ?
Thanks
http://archive.ics.uci.edu/ml/datasets/Plants
with min-conf = 0.92. But with apriori and FP-Growth have difference number of rules. With Apriori it have 212 Rules and FP-Growth have 211 Rules.
How is it difference ?
Thanks
Re: Result rules not match Apriori and FP-Growth
Posted by:
webmasterphilfv
Date: August 28, 2014 11:54AM
Normally, they should output the same result.
If they don't output the same result, it could means that:
1) There is a bug
2) The input format of the input file might be incorrect. Some algorithms assumes that each line (transaction) is sorted by lexical order and that an item does not appear twice in the same line. If this is not followed then some unexpected result may happen.
If you send me your file by e-mail (philippe.fv AT gmail.com), I can check it to see what happens.
Best,
If they don't output the same result, it could means that:
1) There is a bug
2) The input format of the input file might be incorrect. Some algorithms assumes that each line (transaction) is sorted by lexical order and that an item does not appear twice in the same line. If this is not followed then some unexpected result may happen.
If you send me your file by e-mail (philippe.fv AT gmail.com), I can check it to see what happens.
Best,