wuqixiaobai / mtranse

Code and data for IJCAI-17 paper Multilingual Knowledge Graph Embeddings for Cross-lingual Knowledge Alignment

Multilingual Knowledge Graph Embeddings for Cross-lingual Knowledge Alignment

This repository includes the code of MTransE var4 (see paper), links to the data sets, and pretrained models.

Install

Make sure your local environment has the following installed:

Python >= 2.7.6
pip

Install the dependents using:

./install.sh

Run the experiments

Please first download the data sets:

http://yellowstone.cs.ucla.edu/~muhao/MTransE/data.zip

and pretrained models

http://yellowstone.cs.ucla.edu/~muhao/MTransE/models.zip

Unpack these two folders to the local clone of the repository.

To run the experiments on WK3l (wikipedia graphs), use:

./run_wk3l.sh

To run the experiments on CN3l (conceptNet), use:

./run_cn3l.sh

You may also train your own models on these two data sets using:

./train_models.sh

Reference

Please refer to our paper. Muhao Chen, Yingtao Tian, Mohan Yang, Carlo Zaniolo. Multilingual Knowledge Graph Embeddings for Cross-lingual Knowledge Alignment. In Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI), 2017

@inproceedings{chen2017multigraph,
    title={Multilingual Knowledge Graph Embeddings for Cross-lingual Knowledge Alignment},
    author={Chen, Muhao and Tian, Yingtao and Yang, Mohan and Zaniolo, Carlo},
    booktitle={Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI)},
    year={2017}
}