| clear query|facets|time |
Search criteria: recommender.
Results from 71 to 80 from
443 (0.418s).
|
|
|
Did you mean:
|
|
Loading phrases to help you
refine your search...
|
|
Re: Understanding the SVD recommender - Mahout - [mail # user]
|
|
...Fire away. On Thu, Jun 3, 2010 at 3:52 AM, Sean Owen wrote: ...
|
|
|
Author: Ted Dunning,
2010-06-03, 15:15
|
|
|
Re: Feedback on provided recommendations - Mahout - [mail # user]
|
|
...Exactly. In informal experiments I have tested these two systems: a) consider all ratings as positive (whether like or dislike), run boolean recommendations and then filter out explicit...
|
|
... matches to negative ratings b) run boolean recommendations on positive and negative separately and output a score that is a linear combination of the two System (a) was much better than system...
|
[+ show more]
[- hide]
| ... the most stake in. With video recommendation, using views that lasted >30seconds instead of all view starts gave much better recommendations. Recommendations are often even lower commitment than... |
|
|
Author: Ted Dunning,
2010-04-28, 19:19
|
|
|
Re: Feedback on provided recommendations - Mahout - [mail # user]
|
|
...Boolean recommendations makes the assumption that either negative feedback is really positive feedback (controversial, but works reasonably) or that negative feedback is not very useful...
|
|
... and memory and it may or may not help you. More practical is to use boolean recommendations for the positive side and then just add a few heuristics for the negative side. Typical heuristics...
|
|
|
Author: Ted Dunning,
2010-04-27, 18:43
|
|
|
Re: Feedback on provided recommendations - Mahout - [mail # user]
|
|
... wind up recommending everything except what the user likes (if the negative ratings overwhelm the positive in spite of the blending parameters) or you can wind up missing some very good stuff...
|
|
|
Author: Ted Dunning,
2010-04-27, 19:53
|
|
|
Re: Top-N recommendations from SVD - Mahout - [mail # user]
|
|
...I would recommend against a mutable object on maintenance grounds. Better is to keep the threshold that a new score must meet and only construct the object on need. That cuts...
|
|
|
Author: Ted Dunning,
2013-03-06, 16:01
|
|
|
Re: Top-N recommendations from SVD - Mahout - [mail # user]
|
|
...Hmm... each users recommendations seems to be about 2.8 x 20M Flops = 60M Flops. You should get about a Gflop per core in Java so this should about 60 ms. You can make this faster...
|
|
... with more cores or by using ATLAS. Are you expecting 3 million unique people every 80 hours? If no, then it is probably more efficient to compute the recommendations on the fly. How many...
|
[+ show more]
[- hide]
| ... recommendations per second are you expecting? If you have 1 million uniques per day (just for grins) and we assume 20,000 s/day to allow for peak loading, you have to do 50 queries per second peak... |
| .... This is quite plausible with bad arrangement of the inner loops. I think that you would have highest performance computing the recommendations for a few thousand items by a few thousand users at a time... |
|
|
Author: Ted Dunning,
2013-03-06, 03:47
|
|
|
Re: Feedback on provided recommendations - Mahout - [mail # user]
|
|
...Even just having this data to benchmark the SVD algorithms would be very interesting. Completely blinding both user id's and item id's should make anonymity safe, especially if time fr...
|
|
|
Author: Ted Dunning,
2010-04-29, 00:17
|
|
|
Re: Feedback on provided recommendations - Mahout - [mail # user]
|
|
...This is a small data set, but it would be of real value for testing of these kinds of problems. 2010/4/28 Tolga Oral ...
|
|
|
Author: Ted Dunning,
2010-04-28, 19:19
|
|
|
Re: Top-N recommendations from SVD - Mahout - [mail # user]
|
|
...That sounds way too long. How is the U matrix stored? What type? On Wed, Mar 6, 2013 at 6:44 AM, Josh Devins wrote: ...
|
|
|
Author: Ted Dunning,
2013-03-06, 16:02
|
|
|
Re: Top-N recommendations from SVD - Mahout - [mail # user]
|
|
...Well, it would definitely not be the for time I counted incorrectly. Anytime I do arithmetic the result should be considered suspect. I do think my numbers are correct, but then ...
|
|
|
Author: Ted Dunning,
2013-03-06, 10:13
|
|
|
|
