turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Community Home
- :
- Product Help
- :
- RapidMiner Studio Forum
- :
- Re: New Logistic Regression Operator: Strange Beha...

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-21-2017 08:47 PM - edited 02-21-2017 08:50 PM

02-21-2017 08:47 PM - edited 02-21-2017 08:50 PM

Typically, in the absence of knowledge about the relative cost of missclassification errors a classifier shoud classify an observation as a member of the "True Class" if Probability(True) > 0.5. That's the behavior of most classifiers in Rapidminer (including W-Logistic).

The new classifier "Logistic Regression" seems to be the exception. This classifier classifies an observation as True if Prob(True) > 0.3 (or in the Rapidminer terminology : if Confidence(True) > 0.3). I'm attaching a process showings this behavior. Just run it. Plot a histogram of Confidence(True) and color it using the variable Prediction(label).

The pic of the histogram is attached to this message too.

9 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 07:47 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 09:11 AM

02-22-2017 09:11 AM

No. I used the new LogisticRegression operator. I didn't even use cross-validation.

The problem seems to be the GeneralizedLinearRegression routine. I exchanged operator (GLM for Logistic Regression) with the right settings (family=binomial, etc) and I get the same behavior.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 09:12 AM - edited 02-22-2017 09:18 AM

02-22-2017 09:12 AM - edited 02-22-2017 09:18 AM

.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 09:31 AM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 11:31 AM

02-22-2017 11:31 AM

That's very curious. Did you try comparing the results of the Weka version of the logistic regression operator?

Brian T., **Lindon Ventures** - www.lindonventures.com

Analytics Consulting by Certified RapidMiner Analysts

Analytics Consulting by Certified RapidMiner Analysts

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 01:59 PM

02-22-2017 01:59 PM

@yyhuang pointed out to me that it might be related to H2O's f1 optimization of binomal data sets for the GLM algo. http://ethen8181.github.io/machine-learning/h2o/h2

Will continue to investigate.

Regards,

T-Bone

Twitter: @neuralmarket

T-Bone

Twitter: @neuralmarket

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 02:04 PM

02-22-2017 02:04 PM

@Telcontar120 I tested this out using the Weka LR and the old Rapidminer SVM LR algo, both give me a label flip at confidence > 0.5 when using a Generate Data operator set to Random Classification.

I think I'm learning toward the internal F1 measure optimization that H20 is doing behind the scenes for binomal labels, but we're looking into this.

Regards,

T-Bone

Twitter: @neuralmarket

T-Bone

Twitter: @neuralmarket

Highlighted
Options

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-22-2017 02:57 PM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-23-2017 07:43 AM