A Nonnegative Latent Factor Model for Large-Scale Sparse Matrices in Recommender Systems via Alternating Direction Method

Computer Science

• Recommender Systems 100%
• Collaborative Filtering 100%
• nonnegative matrix factorization 66%
• Convergence Rate 66%
• Prediction Accuracy 33%
• Slow Convergence 33%
• Real Data Sets 33%
• Fast Convergence 33%
• Big Data 33%
• Real Application 33%
• Representativeness 33%
• Single Feature 33%
• Learning System 33%

Mathematics

• Matrix 100%
• Sparse Matrix 100%
• Latent Factor 100%
• Convergence Rate 50%
• Representativeness 25%
• Filtering Problem 25%
• Real Data 25%
• Nonnegativity 25%
• Big Data 25%

Keyphrases

• Collaborative Filtering-based Recommender Systems 33%