RE: SVM Implementation in Mahout

2014-06-30 Thread Scott Lett, PhD
Yes, we'd like to help.Is this list where people usually post their projects or wishes? Scott -Original Message- From: Ted Dunning [mailto:ted.dunn...@gmail.com] Sent: Sunday, June 29, 2014 11:04 AM To: user@mahout.apache.org Subject: Re: SVM Implementation in Mahout

Re: SVM Implementation in Mahout

2014-06-30 Thread Ted Dunning
@mahout.apache.org Subject: Re: SVM Implementation in Mahout No there is not yet a move to implement this. WOuld you like to start one? On Sun, Jun 29, 2014 at 6:24 AM, Scott Lett, PhD sl...@holisticmath.com wrote: Is there a current project list and/or wish list for Mahout in the new

RE: SVM Implementation in Mahout

2014-06-30 Thread Scott Lett, PhD
Thanks, Ted. -Original Message- From: Ted Dunning [mailto:ted.dunn...@gmail.com] Sent: Monday, June 30, 2014 12:18 PM To: user@mahout.apache.org Subject: Re: SVM Implementation in Mahout No. For development work, the typical process is to have discussions on the dev list and to post

RE: SVM Implementation in Mahout

2014-06-29 Thread Scott Lett, PhD
: SVM Implementation in Mahout Take a look at the new matrix language bindings and DSL [1] that the Mahout community is working on. We would be interested in a state-of-the-art SVM implementation using these techniques. A map-reduce implementation would not be interesting. How did you think

Re: SVM Implementation in Mahout

2014-06-29 Thread Ted Dunning
. Scott -Original Message- From: Ted Dunning [mailto:ted.dunn...@gmail.com] Sent: Friday, June 27, 2014 7:43 PM To: user@mahout.apache.org Subject: Re: SVM Implementation in Mahout Take a look at the new matrix language bindings and DSL [1] that the Mahout community is working

Re: SVM Implementation in Mahout

2014-06-27 Thread Ted Dunning
Take a look at the new matrix language bindings and DSL [1] that the Mahout community is working on. We would be interested in a state-of-the-art SVM implementation using these techniques. A map-reduce implementation would not be interesting. How did you think to implement this? What kind of

SVM Implementation in Mahout

2014-06-25 Thread venkata ramana
Hi Mahout, I would like to implement SVM in mahout. Could you help me how to start SVM? Thanks, Venkat

Re: SVM Implementation for mahout?

2013-12-08 Thread Fernando Santos
On 06.12.2013, at 19:14, Fernando Santos wrote: Hello, Is there any tested SVM implementation for Mahout? Mahout in action says there is a sequential implementation, but Experimental still. I couldn't find this implementation. Thanks -- Fernando Santos

Re: SVM Implementation for mahout?

2013-12-08 Thread Ted Dunning
The problem of correlation of features is clearly present in text, but it is not so clear what the effect will be. For naive bayes this has the effect of making the classifier over confident but it usually still works reasonably well. For logistic regression without regularization it can

Re: SVM Implementation for mahout?

2013-12-08 Thread Fernando Santos
Hello, I think my problem is related to the fact that the dataset is really unbalanced. My 3 classes distribution are 550k, 150k and 70k. And naivebayes make its classification also based on the probability of a class c over all documents. So probably this unbalance is making a big difference.

Re: SVM Implementation for mahout?

2013-12-08 Thread Ted Dunning
On Sun, Dec 8, 2013 at 5:50 PM, Fernando Santos fernandoleandro1...@gmail.com wrote: Actually I had never heard of PCA and LDA. I'll take a look on it. PCA and LDA are probably not quite what you want for Naive Bayes, especially in Mahout. There is an assumption of a sparse binary

Re: SVM Implementation for mahout?

2013-12-07 Thread Fernando Santos
://issues.apache.org/jira/browse/MAHOUT-227 /Manuel On 06.12.2013, at 19:14, Fernando Santos wrote: Hello, Is there any tested SVM implementation for Mahout? Mahout in action says there is a sequential implementation, but Experimental still. I couldn't find this implementation. Thanks

Re: SVM Implementation for mahout?

2013-12-07 Thread Suneel Marthi
, at 19:14, Fernando Santos wrote: Hello, Is there any tested SVM implementation for Mahout? Mahout in action says there is a sequential implementation, but Experimental still. I couldn't find this implementation. Thanks -- Fernando Santos +55 61 8129 8505 -- Manuel

Re: SVM Implementation for mahout?

2013-12-07 Thread Fernando Santos
/jira/browse/MAHOUT-14 https://issues.apache.org/jira/browse/MAHOUT-227 /Manuel On 06.12.2013, at 19:14, Fernando Santos wrote: Hello, Is there any tested SVM implementation for Mahout? Mahout in action says there is a sequential implementation, but Experimental still. I

Re: SVM Implementation for mahout?

2013-12-07 Thread Lucas Fernandes Brunialti
, Is there any tested SVM implementation for Mahout? Mahout in action says there is a sequential implementation, but Experimental still. I couldn't find this implementation. Thanks -- Fernando Santos +55 61 8129 8505 -- Manuel Blechschmidt M.Sc

SVM Implementation for mahout?

2013-12-06 Thread Fernando Santos
Hello, Is there any tested SVM implementation for Mahout? Mahout in action says there is a sequential implementation, but Experimental still. I couldn't find this implementation. Thanks -- Fernando Santos +55 61 8129 8505

Re: SVM Implementation for mahout?

2013-12-06 Thread Manuel Blechschmidt
, Fernando Santos wrote: Hello, Is there any tested SVM implementation for Mahout? Mahout in action says there is a sequential implementation, but Experimental still. I couldn't find this implementation. Thanks -- Fernando Santos +55 61 8129 8505 -- Manuel Blechschmidt M.Sc. IT Systems