[
https://issues.apache.org/jira/browse/MAHOUT-906?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13169485#comment-13169485
]
Sean Owen commented on MAHOUT-906:
----------------------------------
No I think it's as simple as factoring out this section of code in
GenericRecommenderIRStatsEvaluator, that's all:
FastIDSet relevantItemIDs = new FastIDSet(at);
// List some most-preferred items that would count as (most) "relevant"
results
double theRelevanceThreshold = Double.isNaN(relevanceThreshold) ?
computeThreshold(prefs) : relevanceThreshold;
prefs.sortByValueReversed();
for (int i = 0; i < size && relevantItemIDs.size() < at; i++) {
if (prefs.getValue(i) >= theRelevanceThreshold) {
relevantItemIDs.add(prefs.getItemID(i));
}
}
I had thought you were not changing the estimated-based test? but there it is a
matter of factoring out what splitOneUsersPrefs() does.
> Allow collaborative filtering evaluators to use custom logic in splitting
> data set
> ----------------------------------------------------------------------------------
>
> Key: MAHOUT-906
> URL: https://issues.apache.org/jira/browse/MAHOUT-906
> Project: Mahout
> Issue Type: Improvement
> Components: Collaborative Filtering
> Affects Versions: 0.5
> Reporter: Anatoliy Kats
> Priority: Minor
> Labels: features
> Original Estimate: 48h
> Remaining Estimate: 48h
>
> I want to start a discussion about factoring out the logic used in splitting
> the data set into training and testing. Here is how things stand: There are
> two independent evaluator based classes:
> AbstractDifferenceRecommenderEvaluator, splits all the preferences randomly
> into a training and testing set. GenericRecommenderIRStatsEvaluator takes
> one user at a time, removes their top AT preferences, and counts how many of
> them the system recommends back.
> I have two use cases that both deal with temporal dynamics. In one case,
> there may be expired items that can be used for building a training model,
> but not a test model. In the other, I may want to simulate the behavior of a
> real system by building a preference matrix on days 1-k, and testing on the
> ratings the user generated on the day k+1. In this case, it's not items, but
> preferences(user, item, rating triplets) which may belong only to the
> training set. Before we discuss appropriate design, are there any other use
> cases we need to keep in mind?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
