Speaking from experience, I think that expressing the threshold as a
confidence has some attraction, but can be a bit of a difficult interface.
For instance, the equivalent of a 5 standard deviation threshold of either
0.999 or 0.001 (or did I get those right? Can you tell?). In
either c
[
https://issues.apache.org/jira/browse/MAHOUT-1601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14090078#comment-14090078
]
Harish Kayarohanam commented on MAHOUT-1601:
I am doing analytics for github
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089966#comment-14089966
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user pferrel commented on the pul
if exploration and bootstrap are concerns, in my case its saturation is
achieved by a different methodology. I want this threshold to be, (1) of
course optional, and (2) be expressed in confidence level, %, just to
understand the ballpark in each case.
Ok i think i understand the code to convert c
Please!
On Thu, Aug 7, 2014 at 11:52 AM, Frank Rosner (JIRA)
wrote:
>
> [
> https://issues.apache.org/jira/browse/MAHOUT-1601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089600#comment-14089600
> ]
>
> Frank Rosner commented on MAHOUT-1601:
> -
Yes. This is a good thresholding to do.
Typically I have done this by simply providing a threshold on the LLR score
itself. It is convenient to restate the score itself as the signed square
root of the score since that lets you add information about whether the
cooccurrence is more or less commo
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089760#comment-14089760
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user dlyubimov commented on the p
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089745#comment-14089745
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user dlyubimov commented on the p
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089741#comment-14089741
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user pferrel commented on the pul
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089737#comment-14089737
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user pferrel commented on the pul
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089713#comment-14089713
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user dlyubimov commented on the p
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089649#comment-14089649
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user dlyubimov commented on the p
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089645#comment-14089645
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user dlyubimov commented on the p
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089638#comment-14089638
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user pferrel commented on the pul
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089609#comment-14089609
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user dlyubimov commented on the p
On Thu, Aug 7, 2014 at 11:34 AM, Ted Dunning wrote:
>
>
> Can you say a bit more about what you are trying to do?
>
Thank you.
I would like to customize co-oc code not to just yank top N scored
co-occurrences, but also make sure that all of them satisfy rejection of
coincidence with a given con
[
https://issues.apache.org/jira/browse/MAHOUT-1603?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089599#comment-14089599
]
ASF GitHub Bot commented on MAHOUT-1603:
Github user pferrel commented on the pul
[
https://issues.apache.org/jira/browse/MAHOUT-1601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089600#comment-14089600
]
Frank Rosner commented on MAHOUT-1601:
--
Ok. I will try to add some unit tests for th
[
https://issues.apache.org/jira/browse/MAHOUT-1593?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089585#comment-14089585
]
ASF GitHub Bot commented on MAHOUT-1593:
Github user FRosner commented on the pul
On Wed, Aug 6, 2014 at 5:07 PM, Dmitriy Lyubimov wrote:
> On Wed, Aug 6, 2014 at 5:04 PM, Ted Dunning wrote:
>
> > On Wed, Aug 6, 2014 at 6:01 PM, Dmitriy Lyubimov
> > wrote:
> >
> > > > LLR is a classic test.
> > >
> > >
> > > What i meant here it doesn't produce a p-value. or does it?
> > >
>
[
https://issues.apache.org/jira/browse/MAHOUT-1601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089450#comment-14089450
]
Harish Kayarohanam commented on MAHOUT-1601:
I was digging into to find which
Thinking a bit more about the use of LLR only for similarity. Imagine the case
where you are doing text analysis and have TF-IDF weights in the input matrix.
LLR has one trait that makes me wonder about settling on it alone for general
similarity and it’s more an observation since I have no data
[
https://issues.apache.org/jira/browse/MAHOUT-1601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089344#comment-14089344
]
Frank Rosner commented on MAHOUT-1601:
--
I added some JavaDoc to DummySimilarity. It
[
https://issues.apache.org/jira/browse/MAHOUT-1601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089330#comment-14089330
]
Frank Rosner commented on MAHOUT-1601:
--
I will add some javadoc, if you are OK with
[
https://issues.apache.org/jira/browse/MAHOUT-1605?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Frank Rosner updated MAHOUT-1605:
-
Status: Patch Available (was: Open)
> Make VisualizerTest locale independent
> --
[
https://issues.apache.org/jira/browse/MAHOUT-1605?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089277#comment-14089277
]
ASF GitHub Bot commented on MAHOUT-1605:
GitHub user FRosner opened a pull reques
[
https://issues.apache.org/jira/browse/MAHOUT-1605?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Frank Rosner updated MAHOUT-1605:
-
Description:
h5. Problem
When trying to build Mahout on a machine with a locale that uses a dif
Frank Rosner created MAHOUT-1605:
Summary: Make VisualizerTest locale independent
Key: MAHOUT-1605
URL: https://issues.apache.org/jira/browse/MAHOUT-1605
Project: Mahout
Issue Type: Test
[
https://issues.apache.org/jira/browse/MAHOUT-1600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089109#comment-14089109
]
Nagamallikarjuna commented on MAHOUT-1600:
--
Hi Ted,
I have a plan for adding tw
29 matches
Mail list logo