.png){kind=icon}
.png){kind=icon}
Return
Multiple Feature Similarities based Heterogeneous Graph Representation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.knosys.2025.115232.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Propose a transformation of a heterogenous graph into multiple homogenous sub-graphs based on feature similarities. • Propose a representation method based on multiple feature similarity, called MFS, to learn the representation with both the semantic and structural information in the heterogeneous graph. • Verify via public datasets that MFS outperforms most of the state of the art baseline methods.