Publishing Partner: Cambridge University Press CUP Extra Publisher Login
amazon logo
More Info


New from Oxford University Press!

ad

Vowel Length From Latin to Romance

By Michele Loporcaro

This book "draws on extensive empirical data, including from lesser known varieties" and "puts forward a new account of a well-known diachronic phenomenon."


New from Cambridge University Press!

ad

Letter Writing and Language Change

Edited By Anita Auer, Daniel Schreier, and Richard J. Watts

This book "challenges the assumption that there is only one 'legitimate' and homogenous form of English or of any other language" and "supports the view of different/alternative histories of the English language and will appeal to readers who are skeptical of 'standard' language ideology."


Academic Paper


Title: Building Instance Knowledge Network for Word Sense Disambiguation
Paper URL: http://www.ict.swin.edu.au/personal/shu/acsc2011.pdf
Author: Shangfeng Hu
Email: click here TO access email
Author: Chengfei Liu
Email: click here TO access email
Institution: Swinburne University of Technology
Author: Xiaohui Zhao
Email: click here TO access email
Institution: Eindhoven University of Technology
Author: Marek Kowalkiewicz
Email: click here TO access email
Institution: SAP Research
Linguistic Field: Computational Linguistics; Semantics
Subject Language: English
Abstract: In this paper, a new high precision focused WSD approach is proposed, which not only attempts to identify the proper sense for a word but also provides the probabilistic/L/evaluation for the identification confidence at the same time. A novel Instance Knowledge Network (IKN) is built to generate and maintain semantic knowledge at the word, type synonym set and instance levels. Related algorithms based on graph matching are developed to train IKN with probabilistic knowledge and to use IKN for probabilistic word sense disambiguation. Based on the Senseval-3 all-words task, we run extensive experiments to show the performance enhancements in different precision ranges and the rationality of probabilistic based automatic confidence evaluation of disambiguation. We combine our WSD algorithm with five best WSD algorithms in senseval-3 all words tasks. The results show that the combined algorithms all outperform the corresponding/L/algorithms.
Type: Individual Paper
Status: Completed
Venue: 2011 Australasian Computer Science Conference (ACSC 2011)
URL: http://www.ict.swin.edu.au/personal/shu/acsc2011.pdf


Add a new paper
Return to Academic Papers main page
Return to Directory of Linguists main page