clear query| facets| time Search criteria: author:"Sebastian Schelter".   Results from 21 to 30 from 999 (0.0s).
Loading phrases to help you
refine your search...
lambda overfitting param and ParallelALSFactorizationJob -- suggested value? - Mahout - [mail # user]
...The right value for lambda depends on the data and the confidence function and should be chosen via cross-validation.  Coincidentally, I'm currently watching lecture X (10) of http://ml...
   Author: Sebastian Schelter , 2011-11-18, 16:27
problem:when "no recommendation" ? - Mahout - [mail # user]
...A very simple solution is to present a list of very popular/recent items.  --sebastian  2012/2/24 manish dunani : > hi, > what to do when "no recommendation " comes ? > If...
   Author: Sebastian Schelter , 2012-02-25, 07:43
[expand - 1 more] - RowSimilarityjob - Mahout - [mail # user]
...+user -dev  The input files need to be SequenceFile. RowSimilarityJob is intended to become a method invokable on DistributedRowMatrix as soon as that is ported to the new hadoop api. &...
   Author: Sebastian Schelter , 2010-10-29, 07:44
Mahout usage - Mahout - [mail # user]
...I can only confirm what Sean said, I have also seen companies use Mahout not willing to talk about it publicly.  However I became aware of an interesting usecase via twitter recently, t...
   Author: Sebastian Schelter , 2010-10-01, 15:44
issue about large number of items to recommend - Mahout - [mail # user]
...> It would be impractical for the recommender > to predict a rating on every single items before ranking them.  In the standard item-based approach only items that are similar to ...
   Author: Sebastian Schelter , 2011-11-21, 16:57
[expand - 1 more] - Mahout fpg - Mahout - [mail # user]
...You can use ItemSimilarityJob to find sets of items that cooccur together in your users interactions.  --sebastian   On 20.11.2013 08:11, Sameer Tilak wrote: >  >  ...
   Author: Sebastian Schelter , 2013-11-20, 07:22
Parameter choice and tuning parallelALS - Mahout - [mail # user]
...Hi Pat,  ParallelALSFactorizationJob actually implements two different flavours of matrix factorization, one that is aimed at explicit feedback data (such as ratings):  "Large-scal...
   Author: Sebastian Schelter , 2013-01-02, 16:46
[expand - 1 more] - Mahout compatibility with Hadoop - Mahout - [mail # user]
...Hi Oren,  you should be able to run Mahout with the latest 0.20.2x and 1.0.x and 1.1.x Hadoop releases.  I personally use the current trunk (which will be released as 0.8 soon) wit...
   Author: Sebastian Schelter , 2013-06-17, 13:45
[expand - 1 more] - "Powered By" logo - Mahout - [mail # user]
...Do we have a (legally compliant) "powered by" logo that external users  can put on their website?  --sebastian...
   Author: Sebastian Schelter , 2011-05-13, 13:45
output of Recommender - Mahout - [mail # user]
...The evaluation randomly splits the dataset into training and test set. You see different results for different splits.  --sebastian  On 27.11.2011 15:30, bish maten wrote: > the...
   Author: Sebastian Schelter , 2011-11-27, 15:28