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
High utility pattern mining
Posted by:
john
Date: June 06, 2019 09:42PM
Respected sir,
Iam using spmf open source library,and while running programs, iam able to only run "foodmart.txt" data set. Iam not able to run "retail_utility_spmf.txt" and other similar huim data sets. What could be the reason. thanks for the help
Iam using spmf open source library,and while running programs, iam able to only run "foodmart.txt" data set. Iam not able to run "retail_utility_spmf.txt" and other similar huim data sets. What could be the reason. thanks for the help
Re: High utility pattern mining
Posted by:
webmasterphilfv
Date: June 10, 2019 05:59AM
Dear John,
Thanks for using SPMF. The problem is likely how the "minutil" parameter was set. Generally, as this parameter is set lower, the search space can exponentially increase, as well as the number of patterns and thus the runtime.
Thus, generally, I recommend to start with a very high minutil value and then to decrease it until you find enough patterns.
If you look at the high utility mining papers, for example, we may use "minutil" values such as 1 million or more depending on the dataset.
Hope this helps.
Best regards,
Philippe
Thanks for using SPMF. The problem is likely how the "minutil" parameter was set. Generally, as this parameter is set lower, the search space can exponentially increase, as well as the number of patterns and thus the runtime.
Thus, generally, I recommend to start with a very high minutil value and then to decrease it until you find enough patterns.
If you look at the high utility mining papers, for example, we may use "minutil" values such as 1 million or more depending on the dataset.
Hope this helps.
Best regards,
Philippe