期刊
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
卷 18, 期 3, 页码 1458-1467出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TII.2021.3091435
关键词
Graph neural networks; Logic gates; Informatics; Task analysis; Smoothing methods; Recurrent neural networks; Markov processes; Graph neural network; hybrid-order; oversmoothing; session-based recommendation
类别
资金
- NSFC [61876193]
- Natural Science Foundation of Guangdong Province [2020A1515110337]
- Guangdong Province Key Laboratory of Computational Science, Sun Yat-sen University [2020B1212060032]
- Open Foundation of Guangdong Provincial Key Laboratory of Public Finance and Taxation with Big Data Application
This article proposes a hybrid-order gated GNN (HGNN) for addressing the oversmoothing problem of graph neural networks (GNNs) in session-based recommendation. The HGNN model utilizes hybrid-order propagation and attention mechanism to capture complex dependencies, leading to better performance in experimental evaluations compared to other methods.
Considering sessions as directed subgraphs, graph neural networks (GNNs) are supposed to be capable of capturing the complex dependencies among items and suitable for session-based recommendation. However, deep GNNs suffer from the oversmoothing problem of making all nodes converge to the same value. In session-based recommendation, the subgraphs transformed by short sessions are usually simple, which cause worse oversmoothing problem. To apply GNNs to session-based recommendation sufficiently, in this article, we propose a hybrid-order gated GNN (HGNN) on account of the oversmoothing problem. The proposed HGNN model is based on the hybrid-order propagation, which avoids insignificant patterns and captures complex dependencies in propagation. What's more, the attention mechanism is utilized to learn different weights of orders in propagation. Then, HGNN is applied to session-based recommendation, which results in a new method called SR-HGNN. Experimental results show that SR-HGNN outperforms the state-of-the-art session-based recommendation methods and eases the oversmoothing problem.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据