Natural Language Generation (Level 10) (U04177)

? Credit Points : 10 ? SCQF Level : 10 ? Acronym : INF-4-NLG-4

The area of study called natural language generation (NLG) investigates how computer programs can be made to produce high-quality natural language text or speech from computer-internal representations of information (or other texts). Motivations for this study range from foundational attempts to understand how people produce text and speech (linguistic, psycholinguistic) to entirely practical efforts to produce natural language output for a wide range of applications, including automatic explanation from advisory systems, automatic summarisation from single or multiple documents, machine translation, dialogue systems, human-robot interaction, tutorial systems, and many more.

An introduction to the theory and practice of computational approaches to natural language generation. The course will cover common approaches to content selection and organization, sentence planning, and realisation. The course will cover both symbolic approaches to generation, as well as more recent statistical and trainable techniques. It also aims to provide: An understanding of key aspects of human language production; An understanding of evaluation methods used in this field; Exposure to techniques and tools used to develop practical systems that can communicate with users; Insight into open research problems in applications of natural language generation, e.g., summarization, paraphrase, dialogue, multimodal discourse.

Entry Requirements

? Pre-requisites : Informatics 2A Foundations of Natural Language Processing Introductory Applied Machine Learning

? Prohibited combinations : Natural Language Generation (Level 11)

Subject Areas

Home subject area

Computational Linguistics, (School of Informatics, Schedule O)

Other subject areas

Cognitive Science, (School of Informatics, Schedule O)

Artificial Intelligence, (School of Informatics, Schedule O)

Delivery Information

? Normal year taken : 4th year

? Delivery Period : Not being delivered

? Contact Teaching Time : 2 hour(s) per week for 10 weeks

All of the following classes

Type	Day	Start	End	Area
Lecture	Monday	09:00	09:50	Central

Summary of Intended Learning Outcomes

Given an NLG system students should be able to:
- Provide a written analysis of how the main theories and algorithms behind NLG systems have been incorporated into the system, including an exposition of the theories and algorithms.
- Provide the basis for the evaluation of the system by diagnosing its relations to other NLG systems, and to human performance data.

Given a simple NLG problem, students should be able to use computational tools and methodologies to solve it.

Given a current area of NLG research, students should be able to locate and summarise recent progress in the area.