The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
"[SOLVED] Attribute weighting for unbalanced data"
Contributor II
Hi,
I am trying to experimet with various techniques for attribute weighting for my dataset which is quite unbalanced. I am subsampling the majority class when I am training classifier. My question is, when I am applying "Weight by ..." operators, should I apply them for original (unbalanced) dataset, or for balanced dataset? Intuitively for balanced I'm just not sure.
Thank you.
I am trying to experimet with various techniques for attribute weighting for my dataset which is quite unbalanced. I am subsampling the majority class when I am training classifier. My question is, when I am applying "Weight by ..." operators, should I apply them for original (unbalanced) dataset, or for balanced dataset? Intuitively for balanced I'm just not sure.
Thank you.
Tagged:
0
Answers
yes, in general you should use the balanced data set for any kind of data mining and data analysis, at least as long as you are in the training process.
Performance measurements can also be taken on the original class distributions, depending on the desired output and interpretation of the performance values.
Best regards,
Marius