Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Protective outcomes of high-affinity monoclonal antibodies against drug-resistant plague strains.

Frontiers in immunology·2026
Same author

Quality of life in patients with pseudomyxoma peritonei treated with cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy and implications for ERAS-informed rehabilitation.

European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology·2026
Same author

Modulating proton-coupled electron transfer in 2D metal-covalent organic frameworks for garnering meliorated dual-function photocatalysis.

Journal of colloid and interface science·2026
Same author

Mass Spectrometry Imaging of Glycosphingolipid C═C Position Isomers by On-tissue Capture-Epoxidation on a Diselenide-Functionalized Covalent Organic Framework.

Analytical chemistry·2026
Same author

Embryonic stem cell-derived extracellular vesicles alleviate ovarian dysfunction.

Mechanisms of ageing and development·2026
Same author

Brucellosis Complicated by Thyroid Abscess and Life-Threatening Hemophagocytic Syndrome: A Case Report.

Infection and drug resistance·2026

相关实验视频

Updated: Jun 4, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.6K

局部可解释的垃圾邮件检测模型,具有多头图形通道注意力网络.

Fuzhi Zhang1, Chenghang Huo1, Ru Ma1

  • 1School of Information Science and Engineering, Yanshan University, Qinhuangdao, Hebei Province, China; The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province, Qinhuangdao, Hebei Province, China.

Neural networks : the official journal of the International Neural Network Society
|December 25, 2024
PubMed
概括
此摘要是机器生成的。

这项研究引入了一种新的可解释模型,用于检测欺诈性的在线评论,提高准确性并为垃圾邮件检测提供明确的解释. 新方法通过有效识别恶意行为者,提高了对电子商务平台的信任.

关键词:
在HSIC上,拉索是HSIC的拉索.解释 解释 解释多头图形通道注意力网络多头图形通道注意力网络垃圾邮件发送者检测检测器

更多相关视频

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

367
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

466

相关实验视频

Last Updated: Jun 4, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.6K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

367
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

466

科学领域:

  • 计算社会科学 计算社会科学
  • 人工智能的人工智能
  • 电子商务 安全 安全 电子商务

背景情况:

  • 在线购物严重依赖用户评论,但来自垃圾邮件发送者的欺诈性评论误导了消费者.
  • 现有的垃圾邮件检测方法经常充当黑子,缺乏可解释性.
  • 这种不透明性阻碍了信任和垃圾邮件检测技术的实际应用.

研究的目的:

  • 为在线评论开发一个本地可解释的垃圾邮件发送者检测模型.
  • 解决目前黑子垃圾邮件检测系统缺乏透明度的问题.
  • 为什么评论或用户被标记为垃圾邮件提供可理解的解释.

主要方法:

  • 一个多头图形道注意力网络被设计用于捕捉高阶用户交互.
  • 通过将HSIC拉索算法和随机步行与重启策略相结合,实现了可解释性.
  • 影响检测结果的关键特征被选择为提供最终解释.

主要成果:

  • 拟议的模型在多个基准数据集 (亚马逊,YelpChi,YelpNYC,YelpZip,Yelp_four) 的准确性,精度,回忆和F1测量方面取得了显著的改进.
  • 与最先进的方法相比,具体性能增长在各种指标中从2.9%到16.13%不等.
  • 这种解释方法在无噪声的频率分布和在不同稀疏度水平下的精度方面取得了卓越的性能.

结论:

  • 开发的模型有效地检测垃圾邮件发送者,同时为检测过程提供可解释的见解.
  • 这种可解释的方法提高了电子商务中垃圾邮件检测系统的可靠性和可信度.
  • 这些发现为在线审查分析中更加透明和负责任的AI应用铺平了道路.