Physics

Archaeology

Event (particle physics)

Recommender system

History

Optics

Epistemology

Machine learning

Philosophy

Cognitive psychology

Artificial intelligence

Psychology

Focus (optics)

Data science

Context (archaeology)

Computer science

Quantum mechanics

Mechanism (biology)

Déjà vu

Predicting users’ preferences based on their sequential behaviors in history is challenging and crucial for modern recommender systems. Most existing sequential recommendation algorithms focus on transitional structure among the sequential actions, but largely ignore the temporal and context information, when modeling the influence of a historical event to current prediction.

Déjà vu: A Contextualized Temporal Attention Mechanism for Sequential Recommendation