Javascript must be enabled for the correct page display

Prolog vs XLE: Parsing a fragment of the Romanian language

Tudor, Andreea Ioana (2022) Prolog vs XLE: Parsing a fragment of the Romanian language. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
Bachelor Project Thesis.pdf

Download (853kB) | Preview
[img] Text
toestemming.pdf
Restricted to Registered users only

Download (122kB)

Abstract

Several elementary structures in the Romanian language exhibit complex linguistic phenomena. The clitic doubling phenomenon, among others, has received much interest in the literature not only for its complexity and historical inheritance, but also due to cross-linguistic differences. Even though the theoretical background has been extensively studied, practical solutions that allow an adequate implementation of Romanian direct object constructions are lacking. This study aims to compare a parser for a fragment of the Romanian language built in Prolog using Definite Clause Grammars (DCGs) and a second parser written in Xerox Linguistic Environment (XLE) under the Lexical-Functional Grammar (LFG) formalism. Designing the models requires the development of a complex set of rules that account for clitic doubling and other challenging linguistic features in Romanian. The analysis of some substantial linguistic phenomena is presented along with their implementation in the aforementioned frameworks. Both parsers can identify whether the input is grammatical or not. When it comes to encoding complex linguistic phenomena, Prolog offers a more accessible and adaptive solution for parsing. At the same time, XLE provides an environment that displays more valuable insights for linguistic analysis.

Item Type: Thesis (Bachelor's Thesis)
Supervisor name: Jones, S.M.
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Jul 2022 14:05
Last Modified: 15 Jul 2022 14:05
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/27918

Actions (login required)

View Item View Item