tailieunhanh - Báo cáo khoa học: "It Would Be Much Easier"

The paper proposes a paradigmatic approach to morphological knowledge acquisition. It addresses the problem of learning from examples rules for word-forms analysis and synthesis. These rules, established by generalizing the training data sets, are effectively used by a built-in int e r p r e t e r which acts c o n s e q u e n t l y as a morphological processor within the architecture of a natural language question-answering system. The PARADIGM system has no a priori knowledge which should restrict it to a particular natural language, but instead builds up the morphological rules. | It Would Be Much Easier If WENT Were GOED Dan TUFIS Institute tor Computer Technique and Informatics 8-10 Miciurin Bd. 71316 Bucharest 1 Romania Tel. 653390 Telex 11891-icpci-r ABSTRACT The paper proposes a paradigmatic approach to morphological knowledge acquisition. It addresses the problem of learning from examples rules for word-forms analysis and synthesis. These rules established by generalizing the training data sets are effectively used by a built-in interpreter which acts consequently as a morphological processor within the architecture of a natural language question-answering system. The PARADIGM system has no a priori knowledge which should restrict it to a particular natural language but Instead builds up the morphological rules based only on the examples provided be they in Romanian English French Russian Slovak and the like. 1. INTRODUCTION For highly Inflexional languages encoding all word forms into a word lexicon declarative morphology approach appears to be a poor solution not only due to a great redundancy which for some languages is prohibitive Jappinen l983 but also with respect to some theoretical aspects as for instance descriptional adequacy Wehrll 1985 . Within an inflexional morphology environment Alshawi 1985 we propose a procedural approach based on automatically acquired flexlo-ning paradigms. The paradigmatic model is Incorporated Ihto an experimental system called PARADIGM which Is Intended to partially replace the acquisition and modelling part of our MORPHO lexicon management system Tufis 1987a incorporated by the IURES environment for building natural language applications . The aim of our work is twofold to obtain a sound linguistic tool for word-forms analysis and synthesis which in case of highly Inflexional languages Is by no means a trivial task and to provide for a psychologically motivated behaviour of such a system In dealing with unknown words. In the following we shall dwell on the technical issues connected to