Word Sense Disambiguation with a Similarity-Smoothed Case Library

A word sense disambiguation system which is going to be used as part of a NLP system needs to be large scale, able to be optimised towards a specific task and above all accurate. This paper describes the knowledge sources used in a disambiguation system able to achieve all three of these criteria. It is a hybrid system combining sub-symbolic, stochastic and rule-based learning. The paper reports the results achieved in Senseval and analyses them to show the systemʹs strengths and weaknesses relative to other similar systems.