Module manager: Faruk Mardan
Email: F.Mardan@leeds.ac.uk
Taught: Semester 2 (Jan to Jun) View Timetable
Year running 2024/25
MODL5029M Principles and Applications of Machine Translation
This module is not approved as an Elective
This module aims to equip students with the knowledge and the ability to use machine translation (MT) to support multilingual information needs. It also introduces the theoretical and practical principles of natural language processing using conventional methods as well as Large Language Models. It complements the core modules in (human) Translation and in Computer-Assisted Translation.
This module aims to build:
1. an advanced understanding of the principles of different types of machine translation models;
2. ability to evaluate machine translation quality with human as well as automatic techniques;
3. a good understanding of the theory and application of Natural Language Processing
On completion of this module, students should be able to:
LO1. Explain the principal architectures and different types of machine translation (MT) and their rationales.
LO2 Conduct critical evaluations of MT systems and MT output from a user's perspective.
LO3. Carry out simple Natural Language Processing tasks with small sample-size data
Skills Learning Outcomes
SO1. Critical thinking
SO2. Skills in using software tools
SO3. Skills in evaluating the outputs produced by machines
This module offers a blend of conceptual knowledge and practical experience in machine translation and quality assessment thereof, including human and automatic approaches. It also builds a theoretical foundation of Natural Language Processing to prepare students for a fast-evolving language service industry thanks to advancements in Large Language Models and Artificial Intelligence. To do this, they need a combination of conceptual knowledge and practical experience.
Classes take the form of a blend of lectures and discussion of Machine Translation architecture and Natural Language Processing, accompanied by practical tasks to practice the use of Machine Translation in different scenarios and processing small amounts of language data. Students will also engage with practical scenarios whereby they conduct human and automatic assessment of Machine Translation quality.
Delivery type | Number | Length hours | Student hours |
---|---|---|---|
Lecture | 5 | 1 | 5 |
Practical | 10 | 1 | 10 |
Seminar | 5 | 1 | 5 |
Private study hours | 130 | ||
Total Contact hours | 20 | ||
Total hours (100hr per 10 credits) | 150 |
- In-class weekly exercises focusing on Machine Translation architecture, quality assessment or Natural Language Processing
- Continuous interaction through Minerva
- Feedback on the mini in-class projects on the application of machine translation in real-life scenarios.
Assessment type | Notes | % of formal assessment |
---|---|---|
Case Study | A Case Study of 1,500 words on critical application and evaluation of MT engines with elements of natural language processing | 100 |
Total percentage (Assessment Coursework) | 100 |
Normally resits will be assessed by the same methodology as the first attempt, unless otherwise stated
The reading list is available from the Library website
Last updated: 3/13/2024
Errors, omissions, failed links etc should be notified to the Catalogue Team