Neural Collaborative Filtering (NCF) is a deep learning approach to collaborative filtering for recommendation systems that aims to learn the complex user-item interaction function using neural networks. NCF addresses the limitations of traditional matrix factorization methods by using non-linear neural networks to model user-item interactions.
Architecture
- Input: one-hot encoder user ID and item ID
- Embedding layer converts sparse one-hot vectors into dense user and item embeddings
- Generalized Matrix Factorization (GMF): element-wise product of user and item embeddings with sigmoid after it
- Multi-Layer Perceptron (MLP): concatenates user and item embeddings and applies multiple hidden layers with ReLU activation function with sigmoid after them
- Neural Matrix Factorization (NeuMF): concatenates the outputs of GMF and MLP, applies a projection layer and sigmoid
The approach uses binary cross-entropy loss for implicit feedback (interaction between the user and the item). Negative sampling: uniform sampling or popularity-based sampling strategies
Advantages
- Captures complex non-linear interactions between users and items
- More expressive than traditional matrix factorization
- Flexible architecture that can incorporate various neural network designs
Disadvantages
- Requires more training data and is more computationally intensive than traditional methods
- May overfit on sparse datasets
Possible improvements
- Pre-train GMF and MLP components separately, use the pre-trained weights to initialize NeuMF
- Neural Graph Collaborative Filtering (NGCF): adds GNN and models high-order connectivity in user-item interaction graphs