From Parser to Query: Automated Parsing of Historical Texts

Host Institution: University of Konstanz
Website: https://www.eventbrite.com/e/from-parser-to-query-automated-parsing-of-historical-texts-tickets-897020972877?aff=oddtdtcreator&lang=en-us&locale=en_US&status=30&view=listing

Dates: 10-Jun-2024 - 11-Jun-2024
Location: Konstanz (held online only)

Focus: This workshop provides training in state-of-the-art methods for creating and querying syntactically parsed corpora of historical texts with a focus on dependency parsing of historical French texts. We will begin the first session with a brief presentation of existing resources in both constituency (Penn, common for English historical corpora) and dependency (UD/Conll) formats and discuss some of the difficulties inherent in creating these resources for historical texts. Participants will then learn how to query existing dependency corpora using the GREW Match online platform.
Minimum Education Level: BA+

Special Qualifications:
Participants should be familiar with the concept of a syntactically parsed corpus and have some experience in using one. From a technical point of view, participants should have a Bash shell (installed by default on Mac and Linux, Windows users must install this manually, e.g. CygWin), a good text editor (e.g. Geany, Sublime Text, Notepad++), and Python 3 installed. Knowledge of programming in Python will be helpful for the more advanced topics but is not essential.

Description:
Who: Dr Mathilde Regnault & Dr Tom Rainsford (University of Stuttgart, DE)

When: June 10th & June 11th, 2-4 pm (Berlin, Rome, Paris time)

Where: Online (Zoom)

What: This workshop provides training in state-of-the-art methods for creating and querying syntactically parsed corpora of historical texts with a focus on dependency parsing of historical French texts. We will begin the first session with a brief presentation of existing resources in both constituency (Penn, common for English historical corpora) and dependency (UD/Conll) formats and discuss some of the difficulties inherent in creating these resources for historical texts. Participants will then learn how to query existing dependency corpora using the GREW Match online platform.

In the second session, participants will parse sample texts using a few parsers, from ready-to-use online servers to locally installed ones. We will discuss the outputs and query the resulting treebanks for linguistic structures, such as verbs’ valency and specific complements.

Participants should be familiar with the concept of a syntactically parsed corpus and have some experience in using one. From a technical point of view, participants should have a Bash shell (installed by default on Mac and Linux, Windows users must install this manually, e.g. CygWin), a good text editor (e.g. Geany, Sublime Text, Notepad++), and Python 3 installed. Knowledge of programming in Python will be helpful for the more advanced topics but is not essential.
How: This workshop introduces a flexible payment model, allowing you to contribute based on your means. We suggest a fee of €5 for BA/MA students, €10 for PhD students, and €20 for staff. However, feel free to select the ticket that suits you best. Workshop recordings will be accessible to everyone at no cost a few months post-event. Participation in the live session ensures the opportunity to pose questions, engage with the material, and support ongoing and upcoming workshops by LingLab (Konstanz). All proceeds, 100%, are dedicated to funding current and future workshops, ensuring they remain freely available to all after the workshop concludes. You can watch all our past methods workshops for free here. If you register but can't attend, you will be sent the workshop recordings on the day of the workshop.

Linguistic Field(s): Historical Linguistics

Tuition: 20.0 (Displayed in) Euros
Tuition Explanation: This workshop introduces a flexible payment model, allowing you to contribute based on your means. We suggest a fee of €5 for BA/MA students, €10 for PhD students, and €20 for staff. However, feel free to select the ticket that suits you best. Workshop recordings will be accessible to everyone at no cost a few months post-event. Participation in the live session ensures the opportunity to pose questions, engage with the material, and support ongoing and upcoming workshops by LingLab (Konstanz). All proceeds, 100%, are dedicated to funding current and future workshops, ensuring they remain freely available to all after the workshop concludes. You can watch all our past methods workshops for free here. If you register but can't attend, you will be sent the workshop recordings on the day of the workshop.

Registration: 03-May-2024 to 10-Jun-2024

Contact Person: Massimiliano Canzi
Email: [email protected]

Apply on the web: https://www.eventbrite.com/e/from-parser-to-query-automated-parsing-of-historical-texts-tickets-897020972877?aff=oddtdtcreator&lang=en-us&locale=en_US&status=30&view=listing

Registration Instructions:
Please register on the Eventbrite website: https://www.eventbrite.com/e/from-parser-to-query-automated-parsing-of-historical-texts-tickets-897020972877?aff=oddtdtcreator&lang=en-us&locale=en_US&status=30&view=listing