README.md

### **Dataset summary**
This is a gold-standard benchmark dataset for sentence alignment, between Sinhala-English-Tamil languages. Data had been crawled from the following news websites. The aligned documents annotated in the dataset NLPC-UOM/document_alignment_dataset-Sinhala-Tamil-English had been considered to annotate the aligned sentences.

Army - https://www.army.lk/<br/>
Hiru -http://www.hirunews.lk<br/>
ITN - https://www.newsfirst.lk<br/>
Newsfirst -https://www.itnnews.lk<br/>

The aligned documents have been manually annotated.

### **Dataset**
The folder structure for each news source is as follows.

```python
si-en<br/>
  |--army<br/>
      |--Sinhala<br/>
      |--English<br/>
      |--army.si-en<br/>
  |--hiru <br/>
      |--Sinhala <br/>
      |--English <br/>
      |--hiru.si-en <br/>
  |--itn <br/>
      |--Sinhala <br/>
      |--English <br/>
      |--itn.si-en <br/>
  |--newsfirst <br/>
      |--Sinhala <br/>
      |--English <br/>
      |--newsfirst.si-en <br/>
ta-en<br/>
si-ta<br/>
```

Sinhala/English/Tamil - contain the aligned documents in the two languages with respect to the news source (army/hiru/itn/newsfirst)<br/>
army.si-en - golden aligned sentence alignment. Each sentence is referenced according to the languageprefix_fileid_sentenceId. <br/>


### **Citation Information**
@article{fernando2022exploiting,<br/>
title={Exploiting bilingual lexicons to improve multilingual embedding-based document and sentence alignment for low-resource languages},<br/>
author={Fernando, Aloka and Ranathunga, Surangika and Sachintha, Dilan and Piyarathna, Lakmali and Rajitha, Charith},<br/>
journal={Knowledge and Information Systems},<br/>
pages={1--42},<br/>
year={2022},<br/>
publisher={Springer}<br/>
}