Parallel Knowledge Enhancement Based Framework for Multi-Behavior Recommendation

By Chang Meng et al

Published on Oct. 21, 2023

Read the original document by opening this link in a new tab.

Table of Contents

1. Introduction
2. Related Work
3. Problem Definition
4. Methodology

Summary

The document presents a Parallel Knowledge Enhancement based Framework (PKEF) for multi-behavior recommendation. It addresses issues such as imbalanced data distribution and negative transfer in multi-task learning. The framework consists of two main components: Parallel Knowledge Fusion (PKF) module and Projection Disentangling Multi-Experts (PME) network. PKF enhances representations of different behaviors to correct biases caused by imbalanced behavioral interactions. PME tackles negative transfer by generating behavior-specific expert information and introducing a projection mechanism for aggregation. The model's effectiveness is validated through experiments on real-world datasets.