en-de
en-es
en-fr
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
Transfer learning in social recommendations
Published on Aug 04, 201113829 Views
Online recommendation systems are becoming more and more popular with the development of web. Data sparseness is a major problem for collaborative filtering (CF) techniques in recommender systems, e
Related categories
Chapter list
Transfer Learning in Social Recommendation Systems00:00
Traditional Machine Learning - 100:03
Traditional Machine Learning - 201:11
A Major Assumption01:30
When distributions are different02:14
When Features are different02:46
Transfer Learning: Source Domains03:33
An overview of various settings of transfer learning - 105:18
An overview of various settings of transfer learning - 206:40
TL Resources06:44
Social Media Can Be Bridges in Transfer Learning07:10
Social Web for Transfer Learning07:46
Annotated PLSA Model for Clustering Z - 108:57
Annotated PLSA Model for Clustering Z - 211:43
“Heterogeneous transfer learning for image classification”11:50
Source Data: Unlabeled Documents12:09
Latent Feature Learning by Collective matrix factorization13:03
Optimization14:25
Heterogeneous Transfer Learning Algorithm14:56
Experiment: # documents15:10
Experiment: # Tagged images16:22
Experiment: Noise16:38
Social Recommendations as Source Data16:50
Recommendation Systems - 117:42
Recommendation Systems - 218:04
Product Recommendation as Link Prediction18:33
Essentials of Collaborative Filtering19:00
Data Sparsity in Collaborative Filtering20:03
Transfer Learning for Collaborative Filtering?20:43
Codebook Transfer21:36
Codebook Construction22:00
Knowledge Sharing via Cluster-Level Rating Matrix23:15
Step 1: Codebook Construction24:26
Step 2: Codebook Transfer24:42
Experimental Setup25:15
Experimental Results (1): Books → Movies25:26
Limitations of Codebook Transfer25:44
Coordinate System Transfer26:59
Our Solution: Coordinate System Transfer27:13
IJCAI 2011 Talk27:43
Transfer by Collective Factorization (IJCAI 2011)28:07
Limitation of CST and CBT28:38
Adaptive Transfer Learning29:05
Adaptive: transfer-all and transfer-none29:16
Social Media (Wiki) as Source Data30:06
Social-behavior Transfer Learning for Recommendation Systems30:54
Wikipedia as the source31:07
User contributors - 133:21
User contributors - 233:30
User contributors - 333:37
Transfer Learning via COEDIT - 133:43
Transfer Learning via COEDIT - 235:25
Experiments35:51
Can COEDIT Improve Recommendation Performance?36:45
How Does the Density Affect Results? - 138:10
How Does the Density Affect Results? - 238:29
How Does the Density Affect Results? - 338:33
Efficiency Test - 138:34
Efficiency Test - 238:50
Summary38:54
Conclusions and Future Work39:36