Matrix factorization techniques for recommender systems
In this thesis we study two basic matrix factorization techniques used in recommender systems, namely batch and stochastic gradient descent. Furthermore, data from Epinions.com, consisting of 40163 users and 139738 items is studied and statistically analyzed into its characteristic classes (i.e. use...
Saved in:
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Language: | en_US |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/11610/18038 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In this thesis we study two basic matrix factorization techniques used in recommender systems, namely batch and stochastic gradient descent. Furthermore, data from Epinions.com, consisting of 40163 users and 139738 items is studied and statistically analyzed into its characteristic classes (i.e. users with only
high ratings, items with only low ratings e.t.c. ). Matrix factorization performance is also examined
and prediction accuracy is associated with user and item classes. |
|---|