en-de
en-es
en-fr
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
System-wide effectiveness of active learning in collaborative filtering
Published on Aug 04, 20114179 Views
The accuracy of a collaborative-filtering system largely depends on two factors: the quality of the recommendation algorithm and the number and quality of the available product ratings. In general,
Related categories
Chapter list
System-Wide Effectiveness of Active Learning in Collaborative Filtering00:00
Introduction00:22
Sparsity of the Data00:45
Active Learning With Collaborative Filtering01:41
Definition of the Active Learning Strategy02:25
Objectives03:24
Some State-of-the-art Strategies04:15
Additional Strategies07:08
Evaluation Methodology09:45
Learning Iteration - 111:02
Learning Iteration - 211:20
Learning Iteration - 311:25
Learning Iteration - 411:47
Evaluation: MAE11:57
Evaluation: NDCG13:06
Evaluation: Precision14:00
Evaluation: Coverage14:27
Partially-Randomized Strategies15:18
Evaluation: MAE and NDCG16:45
Conclusion17:16
Thank you18:43