clear query| facets| time Search criteria: author:"Ted Dunning".   Results from 41 to 50 from 5214 (0.0s).
Loading phrases to help you
refine your search...
[expand - 1 more] - tf-idf + svd + cosine similarity - Mahout - [mail # user]
...I have been intermittently following this point.  Some folks have said that having higher dimensional SVD's should change the distribution of distances.  Actually, that isn't quite...
   Author: Ted Dunning , 2011-06-15, 18:32
[expand - 2 more] - Probabilities in Bayesian classifier - Mahout - [mail # user]
...This is what the term Naive is used in the name.  The scores for this kind of algorithm are 0 to 1 or are logarithms of such a number, but are not at all calibrated probabilities.  ...
   Author: Ted Dunning , 2011-06-15, 18:44
Multidimensional log-likelihood similarity - Mahout - [mail # user]
...Yes.  You can turn the normal item-item relationships around to get this.  What you have is an item x feature matrix.  Normally, one has a user x item matrix in cooccurrence a...
   Author: Ted Dunning , 2013-09-30, 05:17
understanding naive bayes + ngrams - Mahout - [mail # user]
...I think your analysis is correct, but you are also probably correct that having multiple levels at the same time would be preferable.  On Wed, Feb 1, 2012 at 1:05 PM, Stuart Smith  ...
   Author: Ted Dunning , 2012-02-01, 21:08
[expand - 1 more] - LDA - Mahout - [mail # user]
...I don't think that the current LDA could be misused this way, but I wouldn't be surprised if the current variational code could be adapted to do it.  On Sat, Jun 18, 2011 at 1:59 AM, Ia...
   Author: Ted Dunning , 2011-06-18, 17:35
packaging a recommender as a war file - Mahout - [mail # user]
...A separate project like this is a better way to package this in any case.  It is bad practice to have developers modifying Mahout itself in order to build their applications.  Nice...
   Author: Ted Dunning , 2012-03-07, 18:52
experimenting with mahout taste and ontologies - Mahout - [mail # user]
...Business logic like this can be built into the IDRescorer capabilities.  There is a lot of information in the mailing list archive on this kind of thing.  See  http://www.luci...
   Author: Ted Dunning , 2012-03-07, 19:08
[expand - 3 more] - Naive-Bayes work flow - Mahout - [mail # user]
...This is a tiny dataset. Have you considered just trying R?  In fact in terms of just diagnosing the problem it would be good to run a regression in R first.   Sent from my iPhone &...
   Author: Ted Dunning , 2012-02-27, 17:39
[expand - 2 more] - Goals for Mahout 0.7 - Mahout - [mail # user]
...Is this a joke?      new String[] {"-t", INPUT_TABLE, "-m", MAIL_ACCOUNT_ID}  seems better than farting around with lists.  On Thu, Feb 23, 2012 at 2:03 PM, Ioan Eug...
   Author: Ted Dunning , 2012-02-23, 15:53
[expand - 3 more] - About Mahout. - Mahout - [mail # user]
...On Mon, Nov 14, 2011 at 7:46 AM, Gustavo Salazar  wrote:  > Ted, > > And those command line programs are necessary when running which features > of mahout? > > An...
   Author: Ted Dunning , 2011-11-14, 15:50