Use this URL to cite or link to this record in EThOS:
Title: Natural Arabic language text understanding
Author: Al-Khonaizi, Mohammed Taqi
ISNI:       0000 0001 3406 7252
Awarding Body: University of Greenwich
Current Institution: University of Greenwich
Date of Award: 1999
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Thesis embargoed until 16 Mar 2017
Access through Institution:
The most challenging part of natural language understanding is the representation of meaning. The current representation techniques are not sufficient to resolve the ambiguities, especially when the meaning is to be used for interrogation at a later stage. Arabic language represents a challenging field for Natural Language Processing (NLP) because of its rich eloquence and free word order, but at the same time it is a good platform to capture understanding because of its rich computational, morphological and grammar rules. Among different representation techniques, Lexical Functional Grammar (LFG) theory is found to be best suited for this task because of its structural approach. LFG lays down a computational approach towards NLP, especially the constituent and the functional structures, and models the completeness of relationships among the contents of each structure internally, as well as among the structures externally. The introduction of Artificial Intelligence (AI) techniques, such as knowledge representation and inferencing, enhances the capture of meaning by utilising domain specific common sense knowledge embedded in the model of domain of discourse and the linguistic rules that have been captured from the Arabic language grammar. This work has achieved the following results: (i) It is the first attempt to apply the LFG formalism on a full Arabic declarative text that consists of more than one paragraph. (ii) It extends the semantic structure of the LFG theory by incorporating a representation based on the thematic-role frames theory. (iii) It extends to the LFG theory to represent domain specific common sense knowledge. (iv) It automates the production process of the functional and semantic structures. (v) It automates the production process of domain specific common sense knowledge structure, which enhances the understanding ability of the system and resolves most ambiguities in subsequent question-answer sessions.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: P Philology. Linguistics ; QA75 Electronic computers. Computer science ; TK Electrical engineering. Electronics Nuclear engineering Computer software Signal processing Information theory Linguistics