COMP3310
Natural Language Processing
10 credits

Taught Semester 2 View Timetable

Year running 2005/06

Pre-requisites
COMP2340

This module is not approved as an Elective

Objectives
On completion of this module, students should be able to:
* understand theory and terminology of empirical modelling of natural language;
* understand and use algorithms, resources and techniques for implementing and evaluating NLP systems;
* be familiar with some of the main language engineering application areas;
* appreciate why unrestricted natural language processing is still a major research task.

Syllabus
Short introduction to linguistic theory and terminology. Algorithms and techniques for computer-assisted text processing; Including algorithms for syntactic, semantic and discourse analysis and covering some of the main problems in these areas (tagging, parsing, word sense disambiguation, anaphora resolution). The focus will be on data-driven, statistic and corpus-based approaches. Evaluation methodologies and problems, linked to human performance in these areas. Resources: Corpus design, annotation and analysis; Machine-Readable dictionaries and lexical databases. Web-based natural language processing. Applications: Current commercial applications for NLP systems and Language Engineering, e.g. machine translation, information extraction.

Teaching methods
Lectures: 20 x 1 hour.

Private study
Taught session follow-up: 20 hours;
Formative assessment: 35 hours;
Summative assessment: 25 hours.

Progress monitoring
2 x assessment as part of the same mini-project.

First assessment: essay not longer than 3 pages.

Second assessment: essay not longer than 7 pages + practical output (e.g., software demonstration or human experiment results depending on the project chosen).

Methods of assessment
Exam: 1 paper, 1 hr 30 mins, 60%, closed book;
Departmental assessment: 40%.

Reading list
The reading list is available from the Library website

Last updated: 21/04/2005

Disclaimer

Undergraduate module catalogue | Taught Postgraduate module catalogue

Errors, omissions, failed links etc should be notified to the Catalogue Team.