Frequently Asked Questions

What is this repository?

It is a library for linguistic data and tools. It presently contains Turkish resources developed by the Computer Engineering and Linguistics departments at Boğaziçi University. You can:

  • Search for data and tools and easily download them.
  • Tools are dockerized, so you can easily access, build, and use them.
  • Authorized users (currently only affiliated with Boğaziçi University) can deposit new resources

What is the PID (handle) good for?

It is a special permanent URL. It provides a permanent link that resolves correctly even if the data is moved elsewhere in the future. Thus, using the URL in citations prevents broken links..

How to cite a resource?

See our policies.

How to get the most out of searches?

In contrast to many search engines, OR is the default operator. Thus, results include resources that include any of the keywords. For more specific searches, the advanced search feature supports searching within specific fields, using negation, emphasizing parts of the query etc. The search engine is is SOLR so use it's syntax if you know it or check it in the documentation.


Named Entity vs Named AND Entity
The default operator is OR. The first example searches for the terms “Named” and “Entity” in all text fields. It will return all resources that contain either or both of these terms. The second example will search for resources that include both terms in any of the fields.
dc.title:Anno* && -dc.title:Web
Returns all items having P?T in title - ? stands for any character (eg. PDT) - and not having WordNet in the title
dc.title:"Named Entity"
Use double quotes (") for exact matches and multiword expressions
author:(Güngör && -Özgür)
Search for items by one author and not the other.

There are many fıelds that are supported. Here is a full record that shows all the supported fields