相关实验视频
在异质信息网络中的链接预测:通过自适应软投票改进了超图卷积
Sheng Zhang1, Yuyuan Huang1, Ziqiang Luo1
1School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, China.
Entropy (Basel, Switzerland)
|February 27, 2026
概括
本研究介绍了VE-HGCN,这是一个用于异质信息网络 (HIN) 中链接预测的新型模型. 它通过将超图卷积与复杂网络分析的软投票合并策略相结合来提高准确性.
科学领域:
- 计算机科学 计算机科学
- 网络分析 网络分析
- 机器学习 机器学习
背景情况:
- 复杂的现实世界系统被建模为异质信息网络 (HIN).
- 传统的链接预测方法在HIN中与高阶结构和语义作斗争.
- 现有的超图模型可能会通过同质化高阶信息来稀释重要的关联.
研究的目的:
- 提出VE-HGCN模型,以改善HIN中的链接预测.
- 解决传统和现有的基于超图的方法的局限性.
- 在复杂的异质网络分析中有效利用高阶信息.
主要方法:
- 从HIN中构建多个异构的超图,使用频繁的子图模式提取.
- 应用超图卷积来进行有效的节点表示学习.
- 采用软投票组合策略来融合多模型预测结果.
主要成果:
- 与七个主流基线模型相比,VE-HGCN模型表现出优越的性能.
- 在四个公开的HIN数据集上进行了实验,验证了模型的有效性.
- 拟议的方法在链接预测准确性方面明显优于现有的方法.
结论:
- VE-HGCN为HIN中的链接预测提供了一个新的视角.
- 该模型对复杂网络分析具有良好的通用性和实用性.
- 这项研究为在HIN中利用高阶信息提供了可行的参考.
相关概念视频
01:07End Point Prediction: Gran Plot
1.3K
A Gran plot is used to predict the equivalence volume or endpoint of a potentiometric or acid-base titration without reaching the endpoint. Typically, titration data is collected as a function of the titrant's volume up to a point less than the equivalence volume and then transformed into a linear format. The straight line is extended to the x-axis, indicating the necessary titrant volume to achieve the equivalence point.
For potentiometric titration, the Gran plot is created by plotting...
For potentiometric titration, the Gran plot is created by plotting...
1.3K
01:03Prediction Intervals
3.5K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
3.5K