# Joint Embedding Learning of Educational Knowledge Graphs

**Authors:** Siyu Yao, Ruijie Wang, Shen Sun, Derui Bu, Jun Liu  
**arXiv:** [1911.08776](https://arxiv.org/abs/1911.08776)  
**Date:** 2019-11-20  
**Summary page:** https://www.summarizepaper.com/en/paper/1911.08776v2/

## Key points

- The paper addresses the problem of embedding learning in educational knowledge graphs.
- Educational knowledge graphs prioritize rich literals over structural relationships.
- The authors propose a novel model that considers both structural and literal information for embedding learning.
- Three experimental graphs based on a real-world educational knowledge graph are constructed.
- The model is evaluated through two experiments on these graphs and other benchmark graphs.
- The experimental results demonstrate the effectiveness of the proposed model for processing educational knowledge graphs.
- The proposed model outperforms other baselines in capturing both structural and literal information.

## Summary

The paper titled "Joint Embedding Learning of Educational Knowledge Graphs" addresses the problem of embedding learning in educational knowledge graphs. While knowledge graphs have been widely used in various fields, such as biomedicine and sociology, their application in education requires a different approach. Unlike other domains where structural relationships are crucial, educational knowledge graphs prioritize rich literals. To tackle this challenge, the authors propose a novel model that considers both structural and literal information to learn embedding representations. They construct three experimental graphs based on a real-world educational knowledge graph used in teaching. The model is evaluated through two experiments on these graphs and other benchmark graphs. The experimental results demonstrate the effectiveness of the proposed model for processing educational knowledge graphs. It outperforms other baselines, highlighting its superiority in capturing both structural and literal information. This research contributes to the advancement of embedding learning techniques specifically tailored for educational knowledge graphs. Authors: Siyu Yao, Ruijie Wang, Shen Sun, Derui Bu, Jun Liu. Title: Joint Embedding Learning of Educational Knowledge Graphs. Abstract: The paper focuses on embedding learning in educational knowledge graphs. While traditional approaches prioritize structural relationships among entities and relations, educational knowledge graphs emphasize rich literals. The authors propose a novel model that incorporates both structural and literal information to learn embedding representations.