Multiword Expressions : Towards Identification to Applications

ISBN-13: 9783659249563 / Angielski / Miękka / 2012 / 296 str.

Tanmoy Chakraborty

Opis

A Multiword Expression (MWE) is a lexeme made up of a sequence of two or more lexemes that has properties which are not predictable from the properties of the individual lexemes or their normal mode of combination. It is a key problem for the development of large-scale, linguistically sound natural language processing technology. An adequate comprehensive analysis of multiword expressions must employ both symbolic and statistical techniques. Beside the identification and extraction, this book covers the usefulness of MWEs in various NLP applications like Machine Translation system, Authorship Identification etc. The whole study has been carried out on English and Bengali languages. Tackling various constraints of acquisition of appropriate resources in NLP technology, all the researches in this book have produced solutions to get rid of the regional language barriers. The book is also the beginning of the emergence of stylometry analysis in Bengali literature along with the parallel references of adaptive techniques in English language.

O autorze

Google India Ph.D fellow in Department of Computer Science, IIT-Kharagpur, India. Received Master degree in Computer Science & Engg. from JU, Kolkata, India and Bachelor degree from WBUT, Kolkata, India. Research interests: Natural Language Processing, Complex Network, Social Network, AI and Information Retrieval.