GSLT: Machine Translation (Level 2)


The aim of this course is to give a research-oriented overview of machine translation focusing on the following areas:

The course is aimed both at students with limited knowledge of the field, and at students with a more extensive background in machine translation, who will be expected to take a more active part in the discussion of current research, and to work on assignments in accordance with their individual background. In this way, the course is meant to contribute to the common platform for students with different backgrounds within GSLT, and also to prepare the students for a specialization in machine translation. In order to follow the course you should have passed the GSLT course NLP (Level 1) or have a comparable competence from elsewhere.

NB: The official language within GSLT is English but we can decide to have the lectures in Swedish instead, provided of course that all the participants are comfortable with this. In any case, participants are free to formulate their contributions to discussions, whether oral or written, in any language that can be understood by the other participants (which in most circumstances means Swedish or English).

The course is divided into 5 parts:

Time Table 2004

 
Week/Date Content Place
38 Lectures GU, Göteborg
39-42 Reading, practical assignments, term paper
43 Lectures GU, Göteborg
44-51, 1-3 Reading, practical assignments, term paper
January 17-18 Closing seminar UU, Uppsala

Deliverables 2004/5

Week/Date Content
October 1 Assignment report 1 on the web
November 15 Assignment report 2 on the web
December 6 Assignment report 3 to Andy Way
January 10 Term project to Anna (UU)

Part 1: Lectures (Göteborg)

Date Time Room Contents Reading Teacher
1 13/9 8-10 C430 Introduction and overview RI (2), J&M (21, 21.1), MK ASH
2 13/9 10-12 C430 Direct translation J&M (21.4) ASH
3 13/9 13-15 C430 Rule-based translation J&M (21.2, 21.3) ASH
4 13/9 15-17 h423 Rule-based translation, Laboratory session Assignment 2 EP
5 20/10 15-17 H421 Reuse of translation 1: Alignment AW AW
6 21/10 8-10 C430 Reuse of translation 2: translation memory, example-based translation, statistical MT AW AW
7 21/10 10-12 C430 Evaluation of MT AW AW

Lecturers: Anna Sågvall Hein (ASH) anna@lingfil.uu.se, Andy Way (AW) away@computing.dcu.ie
Course assistant: Eva Pettersson (EP) evapet@stp.ling.uu.se

Introductory lecture slides will be linked to each topic.

Part 2: Distance Study Periods

The distance learning part of the course consists of five study periods of 2 weeks, 3 weeks, 3 weeks, 3 weeks, and 4 weeks, respectively:

  1. The concept and feasibility of machine translation. Direct translation. [20 September - 1 October] Coordinator: Anna Sågvall Hein
  2. Rule-based translation. [4 October - 25 October] Coordinator: Anna Sågvall Hein
  3. Reuse of translation. [20 October - 15 November] Coordinator: Andy Way
  4. Evaluation. [16 November - 6 December] Coordinator: Andy Way
  5. Focus on term paper [7 December - 10 January] Coordinators: Anna Sågvall Hein, Andy Way
During each of these study periods, students will be engaged in three kinds of activities:

Part 3: Closing Seminar (Uppsala)

The course will end with a seminar, where students will present their papers and the course be informally evaluated. The seminar will be held at Uppsala University, 17 - 18 January, 2005, in room 9-1016.

Course Requirements

In order to pass the course students must fulfil the following requirements:

Reading list, preliminary version

Study period 1 (Human translation. The concept and feasibility of MT. Direct translation)

Reference literature:

Study period 2 (Rule-based translation)

and, maybe, more to come

Study period 3 (Reuse of translation)

The following list is very tentative:

Study period 4 (Evaluation)

The following list is very tentative:
evapet@stp.ling.uu.se
Last update: 1 November 2004