高级检索
汤颖, 周元博, 孙国道. 基于可解释图神经网络的可视推荐分析系统[J]. 计算机辅助设计与图形学学报. DOI: 10.3724/SP.J.1089.null.2023-00549
引用本文: 汤颖, 周元博, 孙国道. 基于可解释图神经网络的可视推荐分析系统[J]. 计算机辅助设计与图形学学报. DOI: 10.3724/SP.J.1089.null.2023-00549
Ying Tang, Yuanbo Zhou, Guodao Sun. A VisualRecommendation Analysis System based on the ExplainingGraphNeural Network[J]. Journal of Computer-Aided Design & Computer Graphics. DOI: 10.3724/SP.J.1089.null.2023-00549
Citation: Ying Tang, Yuanbo Zhou, Guodao Sun. A VisualRecommendation Analysis System based on the ExplainingGraphNeural Network[J]. Journal of Computer-Aided Design & Computer Graphics. DOI: 10.3724/SP.J.1089.null.2023-00549

基于可解释图神经网络的可视推荐分析系统

A VisualRecommendation Analysis System based on the ExplainingGraphNeural Network

  • 摘要: 本文对推荐系统中图神经网络的可解释性进行研究, 从模型出发, 将推荐问题转换为图分类问题, 利用解释图神经网络对推荐系统进行解释, 并突破了以往推荐中解释多为实例级的情况,基于可解释图神经网络提出对推荐模型的从实例级和组群级解释, 支持推荐场景下多粒度的解释. 另外, 为了增强对解释模型提取的图模式的理解, 本文设计了可视分析系统, 来更好地理解图模式和模型解释过程, 从单用户, 用户群, 多个用户群三个层级展开探索, 便于分析人员进行多粒度的图模式挖掘, 并对解释的可靠性进行了可视验证, 帮助分析人员探索图神经网络的推荐模式. 最后, 通过真实的豆瓣电影数据验证了系统的有效性.

     

    Abstract: This paper investigates the interpretability of graph neural networks in recommendation systems by converting the recommendation problem into a graph classification problem and using interpretable graph neural networks to interpret the recommendation system. To enhance understanding of the graph patterns extracted from the interpreted models, this paper designs a visual analysis system, that explores three levels: single user, user group, and multiple user groups. The system also provides a visual verification of the interpretation's reliability and helps analysts explore the recommendation patterns of graph neural networks. Finally, the effectiveness of the system was verified with real Douban movie data.

     

/

返回文章
返回