John Tait, Ph.D.
Professor, Intelligent Information Systems And Associate Dean, School of Computing And Technology, University of Sunderland
Executive Editor of the Cambridge University Press Journal Natural Language Engineering
When & Where
10:00am, 2006-12-01 (Fri.)
Audio-visual Classroom, Department of Library and Information Science, National Taiwan University
Computational Linguistics (CL) has been used for more than 40 years in pursuit of better text information retrieval systems (what we would now call search engines). However applications which show improvements in IR effectiveness have rarely been reported in the literature, despite radical improvements in the quality, coverage and performance of CL systems since the 1960's. The talk will explore why this is the case, at least for the most common retrieval task in which a short keyword query is addressed to a large collection of documents. Of course some more specialised tasks: for example question answering, are critically dependent on CL technology, but there remain many key questions about why CL technology is not more widely adopted. These included
1. Is the conventional information retrieval task formulated in a way which prevents or obstructs computational linguistics contributing ?
2. Does statistical information retrieval in fact capture the relevant properties of language but in a form which is inaccessible or hidden?
3. Are assumptions made in computational linguistics about the nature of lexical semantics and the structural properties of well formed running text in some way ill founded, at least for the information retrieval task?
4. Is there some property of language (for example semantic redundancy) which means that the relatively crude statistical techniques capture enough information to obtain the available improvements in performance?
5. Is the problem that computational linguistic techniques are too unreliable or narrowly applicable, so improved performance on some documents or queries is masked by worse performance on others?
Why is Computational Linguistics not more used in Search Engine Technology?
Speaker: John Tait, Ph.D.
Copyright © 2020,
Theme Originally Created by Devsaran