[
https://issues.apache.org/jira/browse/MAHOUT-420?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12884425#action_12884425
]
Sean Owen commented on MAHOUT-420:
----------------------------------
It's all looking reasonably good. I think the patch may need an update to match
head as I am getting errors applying it. i bet they are small issues.
I skimmed through it and have a few questions:
findDeclaredField() and setField() -- yeah I see what you're doing though it
seems a little fragile to dig inside an object and change its fields. They are
just tests, so maybe it's OK, but are there alternatives? Even for tests, if
it's private, I think it's not testable myself.
Are there any chances to reduce the number of unique writable objects we're
creating? There is some need to specialize and create custom objects for
performance though I do see there are starting to be lots of objects that hold
one or two primitives and I'm keen to reuse classes if reasonable
Likewise I don't mind adding more utility classes per se but I prefer to avoid
utils/helper classes if the methods can be reasonably attached to another
implementation. I haven't looked hard at it, maybe these are necessary, just
noting one concern.
I'll have to look more at the patch when I can apply it and view it in the IDE.
Does this change behavior of the recommender job or is it the same initial
input and final output?
> Improving the distributed item-based recommender
> ------------------------------------------------
>
> Key: MAHOUT-420
> URL: https://issues.apache.org/jira/browse/MAHOUT-420
> Project: Mahout
> Issue Type: Improvement
> Components: Collaborative Filtering
> Reporter: Sebastian Schelter
> Attachments: MAHOUT-420.patch
>
>
> A summary of the discussion on the mailing list:
> Extend the distributed item-based recommender from using only simple
> cooccurrence counts to using the standard computations of an item-based
> recommender as defined in Sarwar et al "Item-Based Collaborative Filtering
> Recommendation Algorithms"
> (http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.144.9927&rep=rep1&type=pdf).
> What the distributed recommender generally does is that it computes the
> prediction values for all users towards all items those users have not rated
> yet. And the computation is done in the following way:
> u = a user
> i = an item not yet rated by u
> N = all items cooccurring with i
> Prediction(u,i) = sum(all n from N: cooccurrences(i,n) * rating(u,n))
> The formula used in the paper which is used by
> GenericItemBasedRecommender.doEstimatePreference(...) too, looks very similar
> to the one above:
> u = a user
> i = an item not yet rated by u
> N = all items similar to i (where similarity is usually computed by
> pairwisely comparing the item-vectors of the user-item matrix)
> Prediction(u,i) = sum(all n from N: similarity(i,n) * rating(u,n)) / sum(all
> n from N: abs(similarity(i,n)))
> There are only 2 differences:
> a) instead of the cooccurrence count, certain similarity measures like
> pearson or cosine can be used
> b) the resulting value is normalized by the sum of the similarities
> To overcome difference a) we would only need to replace the part that
> computes the cooccurrence matrix with the code from ItemSimilarityJob or the
> code introduced in MAHOUT-418, then we could compute arbitrary similarity
> matrices and use them in the same way the cooccurrence matrix is currently
> used. We just need to separate steps up to creating the co-occurrence matrix
> from the rest, which is simple since they're already different MR jobs.
> Regarding difference b) from a first look at the implementation I think it
> should be possible to transfer the necessary similarity matrix entries from
> the PartialMultiplyMapper to the AggregateAndRecommendReducer to be able to
> compute the normalization value in the denominator of the formula. This will
> take a little work, yes, but is still straightforward. It canbe in the
> "common" part of the process, done after the similarity matrix is generated.
> I think work on this issue should wait until MAHOUT-418 is resolved as the
> implementation here depends on how the pairwise similarities will be computed
> in the future.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
