TULAP (Turkish Language Processing Platform) is a software platform that includes natural language processing tools and datasets for Turkish. It has been developed in the scope of a multidisciplinary project by the researchers at Boğaziçi University Computer Engineering and Linguistics departments. This project, led by Suzan Üsküdarlı, Arzucan Özgür, Tunga Güngör and Balkız Öztürk aims to provide open access to the NLP resources developed at our university.
During this project, the team contacted developers of previously developed resources (data and tools) and prepared them for deployment. At the time of the release of TULAP, there were 11 data resources and 14 NLP software tools. We continue to add new resources as they become available. Our students are expected to prepare and deploy their research outcomes on TULAP.
In this project, our main goal was to provide the NLP research community with access to the Turkish NLP resources that we have developed. We specifically focused on preparing openly accessible and well-documented resources. Furthermore, all software tools are dockerized to support ease of use. We believe that easy access to resources will facilitate and speed up further research in these areas.
The following screenshots show the text summarization and named entity recognition tools.
Team:
For more information about the platform and the project team, click here.
All feedback is most welcome.
