provide papers discussing recommender system that use user ratings to measure model's…
- User
provide papers discussing recommender system that use user ratings to measure model's performance
- Assistant
Certainly! Here are some key research papers that discuss recommender systems and use user ratings to measure the model's performance. These papers typically evaluate models using metrics derived from user ratings such as RMSE, MAE, Precision, Recall, or ranking-based metrics.
Classic and Foundational Papers
-
Matrix Factorization Techniques for Recommender Systems
Yehuda Koren, Robert Bell, Chris Volinsky
IEEE Computer, 2009- This seminal paper discusses matrix factorization methods for collaborative filtering and evaluates models using user rating prediction accuracy (RMSE).
- Link (PDF)
-
Collaborative Filtering for Implicit Feedback Datasets
Yifan Hu, Yehuda Koren, Chris Volinsky
2008 IEEE ICDM- Focuses on implicit feedback but also discusses evaluation on explicit user ratings.
- Uses metrics like RMSE and ranking metrics based on user ratings.
- Link (PDF)
Recent and Advanced Papers
-
Neural Collaborative Filtering
Xiangnan He, Lizi Liao, Hanwang Zhang, Liqiang Nie, Xia Hu, Tat-Seng Chua
WWW 2017- Proposes a neural network-based collaborative filtering model.
- Evaluates performance on explicit rating datasets using RMSE and ranking metrics.
- Link (PDF)
-
AutoRec: Autoencoders Meet Collaborative Filtering
Suvash Sedhain, Aditya Krishna Menon, Scott Sanner, Lexing Xie
WWW 2015- Uses autoencoders for rating prediction.
- Evaluates on user rating datasets using RMSE and MAE.
- Link (PDF)
Survey Papers (with evaluation discussions)
-
A Survey of Collaborative Filtering Techniques
M. Su and K. Khoshgoftaar
Advances in Artificial Intelligence, 2009- Reviews collaborative filtering methods and discusses evaluation metrics based on user ratings.
- Link (PDF)
-
Recommender Systems Handbook (Chapter on Evaluation)
Francesco Ricci, Lior Rokach, Bracha Shapira- Comprehensive overview of recommender system evaluation, including rating prediction accuracy metrics.
- Link (Book)
Key Evaluation Metrics Used in These Papers
- RMSE (Root Mean Squared Error): Measures the difference between predicted and actual user ratings.
- MAE (Mean Absolute Error): Average absolute difference between predicted and actual ratings.
- Precision@K, Recall@K: For top-K recommendation lists, often derived from binarized user ratings.
- NDCG (Normalized Discounted Cumulative Gain): Ranking quality metric using user ratings as relevance scores.
If you want, I can also help you find papers focusing on specific types of recommender systems (e.g., deep learning, hybrid, context-aware) or datasets (e.g., MovieLens, Netflix) that use user ratings for evaluation. Just let me know!
-
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.