Fix: NoveltyEvaluator; new EntropyEvaluator #205
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Merge branch '2.0.0' of https://github.com/guoguibing/librec into 2.0.0 # Conflicts: # core/src/main/java/net/librec/data/convertor/TextDataConvertor.java # core/src/main/java/net/librec/eval/ranking/NoveltyEvaluator.java # core/src/main/resources/driver.classes.props
NoveltyEvaluator complete new...
|
Sorry for change list: There are only 2 new classes and the config file. Have had problems to sync my fork. (I am new to github) |
|
@dvelten BTW, we can include your documentation of NMFItemItem into ours(https://www.librec.net/dokuwiki/doku.php). You can email it to me at sunyatong65536@gmail.com after your optimization. Thank you for your contribution ! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The Implementation of NoveltyEvaluator is not that mentioned in the paper. But it is also a well known and often used Evaluator. But it is more a 'Diversity'-'Entropy' - Evaluator only.
I have now created two Implementations. The new corrected NoveltyEvaluator and an EntropyEvaluator containing an modified version of the old NoveltyEvaluator.
Both Implementation use Entropy or Information to do calculations. The new EntropyEvaluator does not take into account if the item purchased is popular or unpopular in respect of probability purchased.
Instead of this the probability space is different.
NoveltyEvaluator: Probability of item purchased
EntropyEvaluator: Probability of item to be in the recommender result list
Both Implementations could be thought to make the same for loops to sum the Entropy of one result list.
But in the EntropyEvaluator you can do an optimization and rewrite the for loop over users into an for loop over items with the pre-calculated item counts (item probabilities).
Best wishes
Daniel