Upload
analu
View
70
Download
1
Tags:
Embed Size (px)
DESCRIPTION
A non-IID Framework for Collaborative Filtering with Restricted Boltzmann Machines. Kostadin Georgiev, VMware Bulgaria Preslav Nakov , Qatar Computing Research Institute ICML, June 17, 2013, Atlanta. 5. Experiments & Evaluation. 1. Overview. - PowerPoint PPT Presentation
Citation preview
Kostadin Georgiev, VMware BulgariaPreslav Nakov, Qatar Computing Research Institute
ICML, June 17, 2013, Atlanta
6. Comparison to Previous Work
1. Overview
A NON-IID FRAMEWORKFOR COLLABORATIVE FILTERING WITH RESTRICTED BOLTZMANN MACHINES
5. Experiments & Evaluation
• We propose a non-IID framework for collaborative filtering
• based on Restricted Boltzmann Machines (RBM)• models both user-user and item-item correlations• uses real values in the visible layer (as opposed to
multinomial) to model user-item ratings, thus taking advantage of the natural order between them
• We further explore the potential of combining the original training data with data generated by the RBM-based model itself in a bootstrapping fashion
• Evaluation: on two MovieLens datasets - with 100K and 1M user-item ratings, respectively
• Results: Rival the state-of-the-art
2. User/Item-based RBM Model
7. Conclusion & Future Work
• Conclusion• proposed a non-IID RBM framework for collaborative filtering• rivals the best previous algorithms, which are more complex
3. Non-IID Hybrid RBM Model
MovieLens 100kMovieLens 1M
• The visible layer• represents either all ratings by a user or all rating for an item• units model ratings as real values (vs. multinomial)• noise-free reconstruction is better
• Remove the IID assumption for the training data• Topology: Unit vij is connected to two independent hidden
layers: one user-based and another item-based.• Missing values (ratings): the generated predictions are used
during testing, but are ignored during training• Training procedure: we average the predictions of the user-
based and of the item-based RBM models
4. Neighborhood + I-RBM
Use the I-RBM predictions from a neighborhood-based algorithm:
However, compute the averages from the original ratings:• Future work• add an additional layer to model higher-order correlations• add content-based features, e.g., demographic
• Two MovieLens datasets:• 100k: • 1,682 movies assigned • 943 users• 100,000 ratings• sparseness: 93.7%
• Each rating is between 1 (worst) and 5 (best)
Data
• 1M:• 3,952 movies• 6,040 users• 1 million ratings• sparseness: 95.8%
- Item-based RBM model outperforms user-based, but not by much.- Hybrid item-based RBM + NB model is relatively insensitive to number of units.
In the IID case, multinomial visible units are better than real-valued. In the non-IID case: real-valued visible units outperform multinomial.
• Mean Absolute Error (MAE):
• Cross-validation• 5-fold• 80%:20% training:testing data splits
Evaluation