Wrong predictions with Graphlab knn collaborative filtering

Asked Apr 18 '18 at 03:40

Active Apr 18 '18 at 03:40

Viewed 64 times

I am using item_similarity_recommender from Graphlab and I noticed that predict() function assigns zero to most of the predicted items even though the ratings go from 1-5. This of course results in having very high RMSE.

Since item_similarity_recommender is based on KNN, I assume that the predicted ratings should be from 1-5. Can anyone explains to me why this is happening??

The code is simply as the following:

train, test = graphlab.recommender.util.random_split_by_user(dataset, max_num_users=1000)
m = graphlab.recommender.item_similarity_recommender.create(train, target='target', only_top_k=65)
rmse=m.evaluate_rmse(test, target='target')
print(m.predict(test))

asked Apr 18 '18 at 03:40

H.H.H

what is target in your case ?, can you out dataset anywhere? – Hari_pb Apr 22 '18 at 06:00
The target is the users' ratings which come in a range from one to five – H.H.H Apr 23 '18 at 15:00
Can you put a snippet of your dataset for reference ?, I am also interested to view target column in snippet. – Hari_pb May 01 '18 at 19:01

Wrong predictions with Graphlab knn collaborative filtering

0 Answers0