LIME

The source code for the paper: On entity alignment at scale.

Credit to the RREA repo.

Dependencies

Python=3.6
Tensorflow-gpu=1.13.1
Scipy
Numpy
Scikit-learn
python-Levenshtein

Datasets

The original datasets are obtained from DBP15K dataset, GCN-Align and JAPE.

Take the dataset DBP15K (ZH-EN) as an example, the folder "zh_en" contains:

ent_ids_1: ids for entities in source KG (ZH);
ent_ids_1_trans_goo: entities in source KG (ZH) with translated names;
ent_ids_2: ids for entities in target KG (EN);
ref_ent_ids: entity links for testing/validation;
sup_ent_ids: entity links for training;
triples_1: relation triples encoded by ids in source KG (ZH);
triples_2: relation triples encoded by ids in target KG (EN);
zh_vectorList.json: the input entity feature matrix initialized by word vectors;

Running

For evaluation on the small and medium-sized datasets without the partition strategies:

Unpackage data.zip
Configure the settings in run.sh.
Then run

bash run.sh

Due to the instability of embedding-based methods, it is acceptable that the results fluctuate a little bit when running code repeatedly.

If you have any questions about reproduction, please feel free to email to zengweixin13@nudt.edu.cn.

Citation

To be added.

Evaluation on the large-scale datasets

For evaluation on the large-scale dataset, open the Large-scale directory. Unpackage the DBP15K.zip, DWY100K.zip. The size of the FB_DBP_2M dataset is too large. You can download it here and put the unzipped files under this directory. Then, run run.py to obtain the results.

Required packages can be found in requirement.txt.