« Return to Thread: Association Rule Generation
Re: Association Rule Generation
by Peter Reutemann-3
::
Rate this Message:
> am new in WEKA and association rule mining. I have the file in arrf fomat I
> can open the file with Weka no problem but I can not get any result it
> thinks and thinks and then generates out of memory error.
How much memory do you allow for the heap?
> I dont have too
> many attributes and all my atributes are nominal. It should not be a
> problem guess. Should the attributes be nominal in order to generate
> association rule.
Apriori handles *only* nominal attributes.
> The only setiing i change is , I set lowerbound to 0.01
> since what I have told.
Normally, if one doesn't get any results with the default settings,
the lowerBoundMinSupport should be slowly lowered, until one gets
results.
Mark also fixed a bug in Apriori today regarding an infinite loop if
the lower bound was too small. Just grab a snapshot from the Weka
homepage, if you want this bugfix.
> I have attached the file could you guys see what is
> wrong with it. What is the meaning of lowerbound.
Copy/paste from Javadoc:
"Iteratively reduces the minimum support until it finds the required
number of rules with the given minimum confidence."
The following three parameters are used for this (once again,
copy/paste from Javadoc):
-D <delta for minimum support>
The delta by which the minimum support is decreased in
each iteration. (default = 0.05)
-U <upper bound for minimum support>
Upper bound for minimum support. (default = 1.0)
-M <lower bound for minimum support>
The lower bound for the minimum support. (default = 0.1)
> I have many null values
> for attributes but each attribute has at least one value. Is there any way
> to ommit null values .
Cheers, Peter
--
Peter Reutemann, Dept. of Computer Science, University of Waikato, NZ
http://www.cs.waikato.ac.nz/~fracpete/ Ph. +64 (7) 858-5174
_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@...
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
> can open the file with Weka no problem but I can not get any result it
> thinks and thinks and then generates out of memory error.
How much memory do you allow for the heap?
> I dont have too
> many attributes and all my atributes are nominal. It should not be a
> problem guess. Should the attributes be nominal in order to generate
> association rule.
Apriori handles *only* nominal attributes.
> The only setiing i change is , I set lowerbound to 0.01
> since what I have told.
Normally, if one doesn't get any results with the default settings,
the lowerBoundMinSupport should be slowly lowered, until one gets
results.
Mark also fixed a bug in Apriori today regarding an infinite loop if
the lower bound was too small. Just grab a snapshot from the Weka
homepage, if you want this bugfix.
> I have attached the file could you guys see what is
> wrong with it. What is the meaning of lowerbound.
Copy/paste from Javadoc:
"Iteratively reduces the minimum support until it finds the required
number of rules with the given minimum confidence."
The following three parameters are used for this (once again,
copy/paste from Javadoc):
-D <delta for minimum support>
The delta by which the minimum support is decreased in
each iteration. (default = 0.05)
-U <upper bound for minimum support>
Upper bound for minimum support. (default = 1.0)
-M <lower bound for minimum support>
The lower bound for the minimum support. (default = 0.1)
> I have many null values
> for attributes but each attribute has at least one value. Is there any way
> to ommit null values .
Cheers, Peter
--
Peter Reutemann, Dept. of Computer Science, University of Waikato, NZ
http://www.cs.waikato.ac.nz/~fracpete/ Ph. +64 (7) 858-5174
_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@...
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
« Return to Thread: Association Rule Generation
| Free embeddable forum powered by Nabble | Forum Help |
