LING 575 F: Topics in Computational Linguistics

Meeting Time: 
Th 3:30pm - 5:50pm
Location: 
SAV 140
SLN: 
16319
Joint Sections: 
LING 575 J, LING 575 G
Instructor: 
Sarah L. Ketchley

Syllabus Description:

URL for Course Survey

https://uw.iasystem.org/survey/191929

You should get an email with a reminder to complete the online course survey. If you don't, please use the link above. The survey is available 5/30/18 through 6/7/18.

Quicklinks

DATA 

GITHUB

ZOOM

READING

 

Welcome to Ling 575: Re-imagining the ‘Golden Age’ of Egyptian Archaeology : Using NLP to Excavate Historical Texts

Thursdays 3.30pm - 5.50pm. Odegaard Video Studio, Room 320, Odegaard Undergraduate Library (Please note room change)

Zoom instructions and information.

Instructor Dr. Sarah Ketchley
Email ketchley@uw.edu
Office Denny M220D
Office Hours Thursdays 1pm - 2.30pm PST or by appointment. In person or via Skype.

This seminar will address the problem of extracting significant information from collections of primary source historical documents of varying quality and content using computational methods. Students will work with travel journals, letters, excavation reports and ephemera from the so-called ‘Golden Age’ of Egyptian archaeology at the end of the 19th and beginning of the 20th centuries. These significant yet understudied documents give the historian a detailed view of the social, geographical and political history of Egypt at the time. Starting with a preliminary collection of journal texts marked up in TEI/XML with a content tag set capturing named entities, students will use computational techniques to scale up work to encode and map a much larger corpus of material. The results of student work will form the basis of data visualizations and analysis of social and historical networks in Egypt and the Near East.

This seminar will offer students a chance to explore how their knowledge of techniques such as named entity recognition, domain adaptation, and sentiment analysis can be applied to support work in digital humanities.  This will involve finding solutions to low-resource scenarios and noisy texts that will likely extend to many other contexts in current applications of NLP.  The seminar will also provide experience in working with collaborators (in this case the instructor, plus possibly other students from other fields) who have domain expertise and NLP-relevant research needs.

Prerequisites

Ling 570 + Ling 571 or 572

General Links and Information

  • Contact, Policies, General Information, Office Hours here.
  • Evaluation and Grading Rubrics here.

Overview of Spring Quarter 2018

This seminar is exploratory in nature, and will evolve based on student data assessments and decisions about team projects and goals. The schedule after week 2 will be updated on a weekly basis to reflect the current status and direction of each team's progress. 

Slides:

Team NTC (Haobo & Yuanhe)

Team OCR (Eric & Jimmy)

Team Pentimento (Audrey, Theresa & Eslam)

 

Week 1

Thursday

March 29th

Welcome & Orientation

Overview

Slides

Week 2

Thursday 

April 5th

Presentations: Data Assessment

Project Planning and Organization

Overview

Slides

Week 3

Thursday 

April 12th

Data selection and project description Remote Group Meetings between 3.30pm - 5.50pm PST
Week 4

Thursday  

April 19th

Team Progress Reports

Overview

 

Week 5

Thursday April 26th

Team Progress Reports

Reversible Transcription Tool: Guest Speakers

Overview 

Week 6 Thursday May 3rd

Team Progress Reports

Newbook Autotagger: Guest Speaker

Overview

 

Week 7 Thursday May 10th Team Progress Reports

Recording 

 

Week 8

Thursday 

May 17th

User Testing

 

 

Week 9

Thursday

May 24th

Team Progress Reports  Recording
Week 10 Wednesday May 31st FINAL CLASS PRESENTATIONS Recording
FINALS WEEK

Thursday June 7th

Final project documentation due in Github No meeting
Catalog Description: 
In-depth study of a particular area of computational linguistics, with hands-on experience. Prerequisite: LING 570 and 571, or permission of instructor. Offered: WSp.
Credits: 
3.0
Status: 
Active
Last updated: 
October 17, 2018 - 9:07pm