Course Title: Natural Language Processing

It is no longer only in Science Fiction that computers are able to understand and produce utterances in natural languages like English and French (either spoken or written). On the other hand, there are still large gaps between computer and human performance on these tasks. This course will review basic techniques for natural language processing, covering applications ranging from spelling correction to translation, information retrieval to speech dialogue systems. The course aims to:

Explain why natural language processing is hard
Give examples of state-of-the-art natural language processing systems
Present useful general concepts and algorithms for building natural language processing systems

A key feature of recent research in natural language processing is the combining together of:

methods based on explicit representation of linguistic knowledge and rule-based processing (the "symbolic approach") and
methods based on numerical analysis of large linguistic corpora (the "statistical approach")

The course will attempt to show the strengths and weaknesses of these two approaches.

The course lasts for 12 weeks and is worth 15 credits. The course organiser is Kees van Deemter. Lecturers are Chris Mellish and Kees van Deemter.

Timetable

Lectures

Monday : 14:00 to 15:00, Meston 3
Tuesday: 13:00 to 14:00, Meston 3

Practicals

Tuesday: 15:00 to 17:00, Meston 311

Practicals start in the second week of teaching.