Recommender Systems

6.

Advanced Recommender Models

6.1.

Context-Aware Recommender Systems

6.1.1.

Defining Context

6.1.1.1.

Temporal Context

6.1.1.2.

Spatial Context

6.1.1.3.

6.1.1.4.

Mood and Activity Context

6.1.1.5.

6.1.2.

Paradigms for Incorporating Context

6.1.2.1.

Contextual Pre-filtering

6.1.2.2.

Contextual Post-filtering

6.1.2.3.

Contextual Modeling

6.1.3.

Context Representation

6.1.3.1.

Contextual Features

6.1.3.2.

Context Vectors

6.2.

Deep Learning for Recommendation

6.2.1.

Neural Network Foundations

6.2.1.1.

Multi-Layer Perceptrons

6.2.1.2.

Embedding Layers

6.2.1.3.

Activation Functions

6.2.2.

Deep Matrix Factorization

6.2.2.1.

Neural Collaborative Filtering

6.2.2.2.

Deep Factorization Machines

6.2.3.

Autoencoders for Recommendation

6.2.3.1.

Denoising Autoencoders

6.2.3.2.

Variational Autoencoders

6.2.4.

Recurrent Neural Networks for Sequential Recommendation

6.2.4.1.

6.2.4.2.

6.2.4.3.

Sequence Modeling

6.2.5.

Graph Neural Networks for Recommendation

6.2.5.1.

Graph Convolutional Networks

6.2.5.2.

Graph Attention Networks

6.2.5.3.

Modeling User-Item Graphs

6.2.6.

Convolutional Neural Networks

6.2.6.1.

CNN for Text and Image Features

6.2.6.2.

Feature Learning

6.3.

Factorization Machines

6.3.1.

Modeling Feature Interactions

6.3.2.

Second-Order Interactions

6.3.3.

Applications in Sparse Data

6.3.4.

Field-Aware Factorization Machines

6.4.

Learning to Rank

6.4.1.

Pointwise Approaches

6.4.1.1.

Regression-Based Methods

6.4.2.

Pairwise Approaches

6.4.2.1.

6.4.2.2.

6.4.3.

Listwise Approaches

6.4.3.1.

6.4.3.2.

6.4.4.

Loss Functions for Ranking

6.4.4.1.

6.4.4.2.

Cross-Entropy Loss

6.5.

Reinforcement Learning for Recommendation

6.5.1.

Multi-Armed Bandits

6.5.2.

Contextual Bandits

6.5.3.

Deep Reinforcement Learning

Previous

5. Hybrid Recommender Systems

Go to top

Next

7. Evaluation of Recommender Systems