| clear query|facets|time |
Search criteria: .
Results from 61 to 70 from
2734 (0.299s).
|
|
|
Loading phrases to help you
refine your search...
|
|
Re: Is Mahout the right tool to recommend cross sales? - Mahout - [mail # user]
|
|
...You can try treating your orders as the 'users'. Then just compute item-item similarities per usual. On Thu, Apr 11, 2013 at 7:59 PM, Billy wrote:...
|
|
|
Author: Sean Owen,
2013-04-11, 19:50
|
|
|
Re: Is Mahout the right tool to recommend cross sales? - Mahout - [mail # user]
|
|
...This sounds like just a most-similar-items problem. That's good news because that's simpler. The only question is how you want to compute item-item similarities. That could be based on user-...
|
|
|
Author: Sean Owen,
2013-04-11, 17:28
|
|
|
Re: Fold-in for ALSWR - Mahout - [mail # user]
|
|
...For simplicity let's consider a brand-new user first, not a new rating for existing user. I'll use the notation from my slides that you mention, A = X * Y'. To clarify, I think you mean you ...
|
|
|
Author: Sean Owen,
2013-04-10, 21:17
|
|
|
Re: log-likelihood ratio value in item similarity calculation - Mahout - [mail # user]
|
|
...Yes using counts is more efficient. Certainly it makes the LLR value different since the results are not normalized; all the input values are N times larger (N = sum k), and so the LLR is N ...
|
|
|
Author: Sean Owen,
2013-04-10, 10:50
|
|
|
Re: log-likelihood ratio value in item similarity calculation - Mahout - [mail # user]
|
|
...These events do sound 'similar'. They occur together about half the time either one of them occurs. You might have many pairs that end up being similar for the same reason, and this is not s...
|
|
|
Author: Sean Owen,
2013-04-10, 09:50
|
|
|
Re: Detecting rank-deficiency, or worse, via QR decomposition - Mahout - [mail # user]
|
|
...PS I think the issue is really more like this, after some more testing. When lambda (overfitting parameter) is high, the X and Y in the factorization A = X*Y' are forced to have a smal...
|
|
|
Author: Sean Owen,
2013-04-08, 08:33
|
|
|
Re: How to map UUID to userId in Preference class to use mahout recommender? - Mahout - [mail # user]
|
|
...You can use the low-order bits, or have a look at what the UUID class does to hash itself to 32 bits in hashCode() and emulate that for 64 bits. Collisions in a 64-bit space are very very ve...
|
|
|
Author: Sean Owen,
2013-04-08, 03:29
|
|
|
Re: Build failed in Jenkins: Mahout-Examples-Cluster-Reuters #290 - Mahout - [mail # dev]
|
|
...No, it builds fine from HEAD for me. This is another transient error from Jenkins. On Sun, Apr 7, 2013 at 6:53 PM, Suneel Marthi wrote:...
|
|
|
Author: Sean Owen,
2013-04-07, 19:20
|
|
|
Re: I believe the TanimotoSimilarity scorer actually uses the Jaccard similarity measure - Mahout - [mail # user]
|
|
...I had not heard of Tanimoto being generalized to n-way similarity, but then again, I can't say I know much at all authoritative about the term. The Wikipedia page says it's incorrectly used ...
|
|
|
Author: Sean Owen,
2013-04-07, 15:09
|
|
|
Re: Detecting rank-deficiency, or worse, via QR decomposition - Mahout - [mail # user]
|
|
...Oh, I thought you were stil referring to the toy example in this thread. Yes, k=10 is definitely larger than its rank. This is why I said Y'*Y is certainly not invertible in this case. I was...
|
|
|
Author: Sean Owen,
2013-04-07, 09:00
|
|
|
|
