English to Persian Wikipedia Translations of Technical Terms

Overview

This repository aims to create a comprehensive dataset that maps technical terms and concepts between English and Persian (Farsi) Wikipedia. The focus is primarily on fields such as Computer Science, Artificial Intelligence, Software Engineering, and Bio-Medical Engineering.

Objectives

• Mapping Technical Terms: Identify and map key technical terms and concepts from English Wikipedia to their corresponding entries in Persian Wikipedia.
• Contextual Information: Include keywords and descriptions to provide context for each term or concept.
• Resource Creation: Develop a resource to aid people in writing seminar papers and theses in Persian, which there is usually strict requirements for using Persian terms. For fields like Computer Science, where there are a ton of English terms and their Persian translations are not always clear, this dataset can relieve some of the pain of finding the right Persian term for a given English term.

Project Structure

en-fa-wikipedia-translations/
│
├── data/
│   ├── dataset.csv
│   ├── dataset.tsv
│   ├── dataset.json
│
├── scripts/
│   ├── query.sparql
│
|── CONTRIBUTING.md
├── .gitignore
├── LICENSE
├── README.md

Accessing The Data

The datasets are readily available in the data/ directory. Furthermore, you can use the SPARQL query in the ./scripts/query.sparql file to fetch the data directly from the Wikidata Query Service.

Dataset Structure

The dataset will include the following fields:

• English Term: The term in English.
• Persian Term: The corresponding term in Persian.
• Description (EN): A brief description of the term in English.
• Description (FA): A brief description of the term in Persian.
• Source URL (EN): The URL of the English Wikipedia entry.
• Source URL (FA): The URL of the Persian Wikipedia entry.

TODO

• Automate the process of fetching the data from the Wikidata Query Service and updating the dataset.
• Optimize the SPARQL query both in terms of performance and the quality of the results.
• Review the mappings to correct any potential inconsistencies or unwanted results.
• Integrate the dataset with other curated datasets for Persian equivalents of technical terms, to create a more comprehensive resource.
• Add workflow to automatically update the dataset with new terms and mappings.

Contributing

Contributions are welcome! To contribute, fork the repository and create a new branch. Make your changes and submit a pull request. For more details, see the CONTRIBUTING file.

License

This project is licensed under the MIT License. See the LICENSE file for more details.

Contact

If you have any questions, suggestions, or feedback, feel free to open an issue or contact me at j.razi@outlook.com.