Title

An Overt Semantics with a Machine-guided Approach for Robust LKBs.

Abstract

In this paper, we report on our experience in building computational semantic lexicons for use in NLP applications. In a machine-guided approach, the computer induces part of the semantic knowledge to be acquired by an acquirer. An overt semantics can help predict the syntactic behavior of words. By overt semantics we mean applying the linking or lexical rules at the semantic level and not on lexical base forms. More specifically, we address the different strategies of acquisition arguing for an application-driven, training-intensive effort. We also report on how to develop lexicons using off the shelf resources, and address multilingual issues. We will try to provide an assessment of the difficulties we encountered and some directions to bypass them.

Comments

7 pages.

Download postscript