School of Computer Science

Module 02495 (2016)

Module description - Natural Language Processing 1

The Module Description is a strict subset of the Syllabus Page.

Module Title Natural Language Processing 1
School School of Computer Science
Module Code 06-02495
Level 2/I
Member of Staff Mark Lee
Semester Semester 1 - 10 credits
Delivery

2 hrs/week lectures and exercise classes.

Contact Hours: 23

Description

The module presents an overview of Natural Language Processing and its applications, followed by introductions to morphology, syntax and semantics. These topics are used to introduce some linguistic theory and appropriate algorithms for their computational implementation. Examples are mostly given using Prolog.

Outcomes

On successful completion of this module, the student should be able to:

  1. Describe major concepts, trends, approaches/systems, and difficulties in Natural Language Processing and the study of language generally.
  2. Discuss and illustrate the potential distinctions between morphology, syntax, semantics and pragmatics
  3. Describe several standard methods of applying morphological and syntactic knowledge in Natural Language Processing systems, for instance: finite-state methods; probabilistic methods; context-free grammars and parsers, including the Active Chart Parsers; unification grammars and parsing; dependency parsing.
  4. Describe the basics of symbolic and statistical semantic algorithms.
  5. Demonstrate knowledge of at least one method for a task such as pronoun reference resolution, coreference resolution, or named-entity recognition as an example of a specific, core task in interpretation.
  6. Describe an application of natural language processing (for instance machine translation or document summarization) and show the place of syntactic, semantic and pragmatic processing.
Assessment

Sessional: 1.5 hr examination (80%), continuous assessment (20%).

Supplementary (where allowed): By examination only.

The nature and timing of the continuous assessment will be specified on the module web page -- see under "Relevant Links".