LING6802
10 units
6000 level
Course handbook
Description
This course offers students the opportunity to work on the problems in existing machine translations. Students will compare multiple machine translation systems in the specific field of their choice (e.g. art, education, engineering, medicine, public health, entertainment), and document the strengths and weaknesses of the evaluated systems in each key linguistic domain: lexicon, morphology and syntax. Students will identify problems in machine translation for a specific field, and develop a strategic system for archiving the errors (e.g. is the ambiguity or lack of naturalness due to the word/lexical choice or the grammatical mismatch?). Students will give feedback on each other’s archiving strategy such that their evaluation procedures improve with peer reviews throughout the course.
Availability
Not currently offered.
This Course was last offered in Semester 2 - 2024.
Learning outcomes
On successful completion of the course students will be able to:
1. Discuss field specific challenges for Machine Translation;
2. Identify field specific goals of human evaluation of Machine Translation performance;
3. Compare the strengths and weaknesses of the existing Machine Translation systems in the selected field;
4. Predict how well Machine Translation systems will function for the selected field;
5. Create a database of evaluation criteria in Machine Translation.
Content
Topics to be covered include the following:
- summary of current Machine Translation performance in a specific field;
- identification of challenges in the specific Machine Translation field;
- setting goals of a team project that evaluates Machine Translation performances in a specific field;
- development of Machine Translation evaluation protocol.
Requisite
To enrol into this course students must have successfully completed LING6801 Introduction to Machine Translation Evaluation.
Assumed knowledge
Students must have a basic knowledge of syntax and morphology, so that they can perform and evaluate the part of speech (POS) tagging of text. Students must also have an advanced comprehension skill in a language other than English.
Assessment items
Report: Report
Proposal / Plan: Plan
Case Study / Problem Based Learning: Case Study
Professional Task: Professional Task
Course outline
Course outline not yet available.
The University of Newcastle acknowledges the traditional custodians of the lands within our footprint areas: Awabakal, Darkinjung, Biripai, Worimi, Wonnarua, and Eora Nations. We also pay respect to the wisdom of our Elders past and present.