Abstract: Session-based recommendations (SBR) play an important role in many real-world applications, such as e-commerce and media streaming.

A-PGNN combines personalized graph neural network and dot-product attention mechanism to learn the transitions on both item level and session level for better ...

Mar 1, 2023 · In this paradigm, co-occurrence pattern can be regarded as a high-order feature embedding for sequential pattern to enhance recommendation.

Oct 24, 2021 · Its purpose is to make recommendations based on the interaction behavior of anonymous users in a short period of time. Graph neural network can ...

A novel sequential dependency enhanced graph neural network (SDE-GNN) is proposed to capture both sequential dependencies and item transition relations over ...

Aiming at this deficiency, we propose a novel sequential dependency enhanced graph neural network (SDE-GNN) to capture both sequential aependencies and item ...

Based on the session graph, GNN can cap- ture complex transitions of items, which are difficult to be revealed by previous conventional sequential methods. Each.

In detail, instead of learning sequential dependency, we propose to adopt pre-training techniques and graph neural network to fully embed user-based ...

ABSTRACT. Session-based recommendations (SBRs) capture items' dependen- cies from the sessions to recommend the next item. In recent years,. Graph neural ...

Missing: Dependency | Show results with:Dependency