YATSI pseudo code

[Valerio jus]

Hi Peter, 

I hope you are well. 

The pseudo code for YATSI algorithm (in the below reference) is indicating the input but it does not indicate the output. Therefore, I would highly appreciate it if you can help by clearly advising on what the output statement needs to be placed in the YATSI pseudo code.

Driessens, K., Reutemann, P., Pfahringer, B., & Leschi, C. (2006, April). Using weighted nearest neighbor to benefit from unlabeled data. In Pacific-Asia Conference on Knowledge Discovery and Data Mining (pp. 60-69). Springer, Berlin, Heidelberg.

Your cooperation is highly appreciated.

Kind regards, 

Valerio

[Peter Reutemann]

> The pseudo code for YATSI algorithm (in the below reference) is indicating the input but it does not indicate the output. Therefore, I would highly appreciate it if you can help by clearly advising on what the output statement needs to be placed in the YATSI pseudo code.

Not quite sure what output you are referring to that is missing in the
pseudo code. All it does is predicting labels, just like any other
classifier that works with nominal classes. It merely tries to
leverage the (larger) unlabeled dataset to improve the model.

From the pseudo code:
Step 1: “pre-label” all the examples
Step 2: Predict the class

That original code/paper was written 16 years ago...

Not sure whether the last release still works in current Weka versions:
https://github.com/fracpete/collective-classification-weka-package

Cheers, Peter
--
Peter Reutemann
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 858-5174 (office)
+64 (7) 577-5304 (home office)
https://www.cs.waikato.ac.nz/~fracpete/
http://www.data-mining.co.nz/

[Valerio jus]

Dear Peter, 

Thank you for the prompt reply. 

Sorry, if I was not clear enough. I used YATSI in my paper where I adopted its pseudo code from your research. As a result, I received the following comment from the reviewer:

"High level pseudo code for the two-stage YATSI algorithm is indicated the input but is not indicated the output."

Kindly what do you suggest as a response to the reviewer?

Your cooperation is highly appreciated.

Thank you once again.

Kind regards, 

Valerio

--
You received this message because you are subscribed to the Google Groups "The ADAMS Flow User mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to theadamsflow-u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/theadamsflow-user/CAHoQ12KwgYL_Z4N5Xw%3DBYmDvZb%3D2OfeKgsPSLye0QJaBcHU%3DKg%40mail.gmail.com.

[Peter Reutemann]

> Sorry, if I was not clear enough. I used YATSI in my paper where I adopted its pseudo code from your research. As a result, I received the following comment from the reviewer:
>
> "High level pseudo code for the two-stage YATSI algorithm is indicated the input but is not indicated the output."
>
> Kindly what do you suggest as a response to the reviewer?

How about:
YATSI predicts classes as per its pseudo-code (in Step 2). The
reference implementation is a meta-classifier within the Weka machine
learning workbench that can be applied to datasets with nominal class
attributes.

[Valerio jus]

Thank you, Peter. 

I have attached an image that reflects the input in pseudo code.

Is there any mathematical language (math symbols) that can be used (similar to the attached input style) to reflect the output? 

Kind regards, 

Valerio

--
You received this message because you are subscribed to the Google Groups "The ADAMS Flow User mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to theadamsflow-u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/theadamsflow-user/CAHoQ12KboBCHjB2tKQjRti3zBUKmDoeenR7dayKxtcPyHsGS%3DQ%40mail.gmail.com.

[Peter Reutemann]

> I have attached an image that reflects the input in pseudo code.
>
> Is there any mathematical language (math symbols) that can be used (similar to the attached input style) to reflect the output?

I'm not an academic.

The reviewer seems to be nitpicking, so just add:

Output: class label.

Please note, in the Weka framework, a classifier can either output an
array with the class probability distribution or the index of the
class label (which corresponds to index in the probability array with
the highest value).

[Valerio jus]

Dear Peter. 

I think you are right; the reviewer seems to be nitpicking :-)

Sure, will consider your advice about writing the output. 

Thank you so much. So happy to talk with you after all this time. 

Take care and have a nice day. 

Kind regards,

Valerio

--
You received this message because you are subscribed to the Google Groups "The ADAMS Flow User mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to theadamsflow-u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/theadamsflow-user/CAHoQ12LYnwnQQN4a1xyp%3D3zjRmCd9b2uC3w3PBrs21d8AkeT%3DA%40mail.gmail.com.