LiteralKG

LiteralKG, a novel Knowledge Graph Embedding Model, specialises in fusing different types of Literal information and entity relations into unified representations, which benefit companion animal disease prediction. LiteralKG is developed by NS Lab, CUK based on pure PyTorch backend.

1. Overview

Over the past few years, Knowledge Graph (KG) embedding has been used to benefit the diagnosis of animal diseases by analyzing electronic medical records (EMRs), such as notes and veterinary records. However, learning representations to capture entities and relations with literal information in KGs is challenging as the KGs show heterogeneous properties and various types of literal information. Meanwhile, the existing methods mainly aim to preserve graph structures surrounding target nodes without considering different types of literals, which could also carry significant information. We propose LiteralKG, a knowledge graph embedding model for the efficient diagnosis of animal diseases, which could learn various types of literal information and graph structure and fuse them into unified representations. We construct a knowledge graph that is built from EMRs along with literal information collected from various animal hospitals. We then fuse different types of entities and node feature information into unified vector representations through gate networks. Finally, we propose a self-supervised learning task to learn graph structure in pretext tasks and then towards various downstream tasks. Experimental results on link prediction tasks demonstrate that our model outperforms the baselines that consist of state-of-the-art models.

The overall architecture of LiteralKG.

2. Reproducibility

Setup environment for running:

• Running: python -m venv LiteralKG

Connect to the environment:

• On window OS run: env.bat
• On linux OS or MAC OS run: source image/bin/activate

Install pip packages:

pip install -r requirements.txt

To build model:

• Running python main.py

There are large attribute features (.pickle files) that can be downloaded at here

3. Reference

:pagewithcurl: Paper * *

:pencil: Blog on Network Science Lab *

4. Citing LiteralKG

Please cite our paper if you find Literal useful in your work: @Article{Hoang2023, author = {Van Thuy Hoang and Thanh Sang Nguyen and Sangmyeong Lee and Jooho Lee and Luong Vuong Nguyen and O-Joun Lee}, title = {Companion Animal Disease Diagnostics Based on Literal-Aware Medical Knowledge Graph Representation Learning}, journal = {IEEE Access}, year = {2023}, volume = {11}, pages = {114238--114249}, month = oct, issn = {2169-3536}, doi = {10.1109/ACCESS.2023.3324046}, }

5. Contributors