| clear query|facets|time |
Search criteria: RecommenderJob.
Results from 81 to 90 from
557 (0.293s).
|
|
|
Loading phrases to help you
refine your search...
|
|
Re: Cluster-based recommenders - Mahout - [mail # user]
|
|
...Be aware that cluster based recommenders almost never perform as well as user/item based recommenders. On Mon, Mar 12, 2012 at 10:03 AM, Ahmed Abdeen Hamed wrote: ...
|
|
|
Author: Ted Dunning,
2012-03-12, 17:49
|
|
|
Re: Item Recommendations - Time based - Mahout - [mail # user]
|
|
...Sean's comment is dead-on and your design inclinations are just fine. Hadoop can (eventually) help with the offline item similarity computation. The existing Mahout recommendation engine...
|
|
... can do the actual item recommendation work at very high speed with an appropriate data store. On Mon, Mar 12, 2012 at 10:25 AM, Mridul Kapoor wrote: ...
|
|
|
Author: Ted Dunning,
2012-03-12, 17:48
|
|
|
Re: Item Recommendations - Time based - Mahout - [mail # user]
|
|
...I would generally recommend using the LLR similarity. But if you have an itch, scratch it. I do think we have a tanimoto similarity already, possibly under a slightly different name...
|
|
|
Author: Ted Dunning,
2012-03-12, 22:59
|
|
|
Re: Recommended reading - Mahout - [mail # user]
|
|
...Gene Golub is generally the one who did lots for LAPACK, but Charles van Loan is pretty awesome as well. The philosophy in the book is all good even if a few recent approaches aren't c...
|
|
|
Author: Ted Dunning,
2011-03-09, 23:27
|
|
|
Re: Recommended reading - Mahout - [mail # user]
|
|
...Golub and van loan talk about parallelization but entirely on the context of shared memory. The cost trade offs are very different in map reduce where iterations are so expensive Sent...
|
|
|
Author: Ted Dunning,
2011-03-09, 06:53
|
|
|
Re: Recommended reading - Mahout - [mail # user]
|
|
...Exactly. Glob and van loan is the standard for computational linear algebra. Sent from my iPhone On Mar 8, 2011, at 7:59 PM, Vipul Pandey wrote: ...
|
|
|
Author: Ted Dunning,
2011-03-09, 06:47
|
|
|
Re: Recommended reading - Mahout - [mail # user]
|
|
...They won't. Strang is the answer for the linear algebra part. On Tue, Mar 8, 2011 at 4:18 PM, Mike Nute wrote: ...
|
|
|
Author: Ted Dunning,
2011-03-09, 00:38
|
|
|
Re: Recommended reading - Mahout - [mail # user]
|
|
...Strang is very hard to beat for linear algebra. For statistics related to machine learning, I would avoid normal statistical texts and go with these instead Pattern Recognition a...
|
|
|
Author: Ted Dunning,
2011-03-08, 20:11
|
|
|
Re: Item Recommendations - Time based - Mahout - [mail # user]
|
|
...Actually I don't think that you will need to implement your own item similarity. Just preprocess your input by grouping by user and sorting by time. Then break user sessions into sepa...
|
|
|
Author: Ted Dunning,
2012-03-12, 18:52
|
|
|
Re: Item Recommendations - Time based - Mahout - [mail # user]
|
|
...In order to get time similarity that you want, you can have virtual users for each session as well as real users for longer time periods. The longer periods will have weaker statistics...
|
|
|
Author: Ted Dunning,
2012-03-12, 17:46
|
|
|
|
