Postgraduate Course: Machine Translation (Level 11) (INFR11062)
|School||School of Informatics
||College||College of Science and Engineering
|Credit level (Normal year taken)||SCQF Level 11 (Year 4 Undergraduate)
||Availability||Available to all students
|Summary||Machine Translation deals with computers translating human languages (for example, from Arabic to English). The field is now sufficiently mature that Google use it to allow millions of people to translate Web Documents each day. This course deals with all aspects of designing, building and evaluating a range of state-of-the-art translation systems. The systems covered are largely statistical and include: word-based, phrase-based, syntax-based and discriminative models. As well as exploring these systems, the course will cover practical aspects such as using very large training sets, evaluation and the open problem of whether linguistics can be useful for translation.
History of MT
* rule-based systems, ALPAC report, IBM models, phrase-based systems
* Factored Models
* Lexicalised reordering
* Changing the source
* Ngram models
* Scaling LMs (cluster-based LMs, Bloom Filter LMs)
* Knight on complexity, problem statement
* Stack decoding
* Human evaluation
* Automatic methods
* NIST competitions
* As factors
Parallel corpora etc (data)
*What they are, where they come from
* Comparable corpora
* Multi-parallel corpora
Relevant QAA Computing Curriculum Sections: Artificial Intelligence, Human-Computer Interaction (HCI), Natural Language Computing
Entry Requirements (not applicable to Visiting Students)
|| It is RECOMMENDED that students have passed
Advanced Natural Language Processing (INFR11059) OR
Accelerated Natural Language Processing (INFR11125) OR
Foundations of Natural Language Processing (INFR09028)
||Other requirements|| This course is open to all Informatics students including those on joint degrees. For external students where this course is not listed in your DPT, please seek special permission from the course organiser.
Advanced Natural Language Processing or equivalent.
Students are expected to understand the following topics, or be prepared to learn them independently.
- Discrete mathematics: analysis of algorithms, dynamic programming, basic graph algorithms, finite and pushdown automata.
- Other essential maths: basic probability theory; basic calculus and linear algebra; ability to read and manipulate mathematical notation including sums, products, log, and exp.
- Programming: ability to read and modify python programs; ability to design and implement a function based on high-level description such as pseudocode or a precise mathematical statement of what the function computes.
- Linguistics: willingness to learn basic elements of linguistic description; no formal linguistics background required.
Information for Visiting Students
|High Demand Course?
Course Delivery Information
|Academic year 2015/16, Available to all students (SV1)
|Learning and Teaching activities (Further Info)
Lecture Hours 20,
Summative Assessment Hours 2,
Programme Level Learning and Teaching Hours 2,
Directed Learning and Independent Learning Hours
|Assessment (Further Info)
|Additional Information (Assessment)
||Coursework dealing with some practical aspect of translation. This will involve analysis of particular problems (for example, reordering or word selection), as well as actual problem solving.
You should expect to spend approximately 40 hours on the coursework for this course.
If delivered in semester 1, this course will have an option for semester 1 only visiting undergraduate students, providing assessment prior to the end of the calendar year.
||Hours & Minutes
|Main Exam Diet S2 (April/May)||Machine Translation||2:00|
On completion of this course, the student will be able to:
- Provide a written description of the main algorithms used in the system.
- Design and justify an approach to the evaluation of the system using state of the art tools and metrics
- Analyse the data collected by such an evaluation.
- Where the system is designed to deal with large volumes of data the student should also be able to describe how the system handles large data volumes and critically compare the system=s solution with other common solutions to the problem.
- Identify where linguistics knowledge is relevant in the design of the system and what influence of linguistic knowledge has on the translation quality and performance of the system.
|* Statistical Machine Translation, P. Koehn, Cambridge University Press, 2010|
* Primary literature
|Course organiser||Dr Adam Lopez
Tel: (0131 6)50 4430
|Course secretary||Ms Sarah Larios
Tel: (0131 6)51 4164
© Copyright 2015 The University of Edinburgh - 18 January 2016 4:13 am