'[Wekalist] computing AUC'

[prev in list] [next in list] [prev in thread] [next in thread] 

List:       wekalist
Subject:    [Wekalist] computing AUC
From:       "Kamphuis, C. (Claudia)" <C.Kamphuis () uu ! nl>
Date:       2008-06-30 13:53:09
Message-ID: 5FE506E2DC924946BB1F0681BBA0881E0152079D () uu01msg-exb04 ! soliscom ! uu ! nl
[Download RAW message or body]

--===============0527012922==
Content-class: urn:content-classes:message
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C8DAB8.A18AC0A9"

This is a multi-part message in MIME format.

Dear all,

I'm interested in computing the area under the curve (AUC). I know that the value for \
the total AUC is computed and visible when running a model in the Explorer and \
selecting 'visualize threshold curve' from the result list.  However...I'm only \
interested in a part of the ROC and to be more specific, that part where the \
False_Positive_Rate is 0.03 at maximum (or where SP is 97% or more).  Is it possible \
to compute a partial area under the curve in WEKA, preferably using either the \
knowlegde flow, the experimenter or the explorer as I don't have any experience in \
using the Simple CLI.  And does anyone know whether the computed AUC is a parametric, \
semi-parametric or non-parametric one?  
Looking forward to some help. Thanks in advance,

Claudia Kamphuis

[Attachment #3 (text/html)]

<HTML dir=ltr><HEAD>
<META http-equiv=Content-Type content="text/html; charset=unicode">
<META content="MSHTML 6.00.2900.3354" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT face=Arial color=#000000 size=2>Dear all,</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>I'm interested in computing the area under the curve \
(AUC). I know that the value for the total AUC is computed and visible when running a \
model in the Explorer and selecting 'visualize threshold curve' from the result list. \
</FONT></DIV> <DIV><FONT face=Arial size=2>However...I'm only interested in a part of \
the ROC and to be more specific, that part where the False_Positive_Rate is 0.03 at \
maximum (or where SP is 97% or more). </FONT></DIV> <DIV><FONT face=Arial size=2>Is \
it possible to compute a partial area under the curve in WEKA, preferably using \
either the knowlegde flow, the experimenter or the explorer as I don't have any \
experience in using the Simple CLI. </FONT></DIV> <DIV><FONT face=Arial size=2>And \
does anyone know whether the computed AUC is a parametric, semi-parametric or \
non-parametric one?</FONT></DIV> <DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>Looking forward to some help. Thanks in \
advance,</FONT></DIV> <DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>Claudia Kamphuis</FONT></DIV>
<DIV><FONT face=Arial color=#000000 size=2></FONT>&nbsp;</DIV></BODY></HTML>

_______________________________________________
Wekalist mailing list
Wekalist@list.scms.waikato.ac.nz
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist

--===============0527012922==--

[prev in list] [next in list] [prev in thread] [next in thread] 