Natural Language Processing, Information Retrieval, Machine Learning and Artificial Intelligence
Supervisor
Suitable for
Abstract
Dr. Clark is happy to supervise projects in Natural Language Processing, Information Retrieval, Machine Learning and Artificial Intelligence.
In particular, Dr. Clark has developed a state-of-the-art natural language parser. A parser for natural language is able to infer the grammatical relations between words in a sentence, indicating roughly who did what to whom. This is the first step towards enabling the computer to "understand" text, and necessary for language processing applications which require some understanding, for example Machine Translation, Dialogue Systems, and Intelligent Search and Retrieval.
The parser uses a sophisticated combination of Linguistic Theory and Machine Learning to accurately and efficiently analyse sentences of English. There are a number of possible projects associated with the parser, especially the Machine Learning component.
The source code for the parser and related software is downloadable from Dr. Clark's webpage (follow the "Software" link).
Prerequisites:
Intelligent Systems I and/or II would be useful.
Programming experience, ideally in C++ (but experience of Java is acceptable).
No understanding of linguistics is required (but an interest in language would help).