1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Grammar-based and example-based translation techniques for handling word agreement and ordering from english to arabic in machine translation
Authors:	Mouiad Fadeil Moh'd Alawneh (P35233)
Supervisor:	Tengku Mohammad Tengku Sembok, Prof. Dr.
Keywords:	Grammar Translation techniques English Arabic Machine translation Natural language processing (Computer science)
Issue Date:	23-Sep-2013
Description:	Machine translation (MT) has been defined as the process that utilizes computer software to translate text from one natural language to another. This definition involves accounting for the grammatical structure of each language and using rules, examples and grammars to transfer the grammatical structure of the source language (SL) into the target language (TL). There is very few MT system that satisfies user needs. While the specific purpose MT systems satisfied a high levels of accuracy, but these systems are restricted in specific domains. Poor performance of MT from English to Arabic has motivated this research to build a MT system where word agreement and ordering are important to ensure the generation of sentences in the TL. Mistakes in MT output can either be the result of analysis at the SL level, or due to generation problems at the target TL level. Some of this agreement should be handled between the subject and verb where the number, gender, person and features of the subject are important factors in verb derivation, as well as verb features. Some other agreements are required between the adjective and the noun where the Arabic adjective depends on the number, gender and person of the noun; other agreements are those between the numbers and the countable noun. This study presents English to Arabic approach for translating well-structured English sentences into well-structured Arabic sentences, using a grammar-based and example-translation techniques to handle the problems of ordering and agreement. This technique combines rule-based MT (RBMT) and example-based MT (EBMT) which is called hybrid-based MT (HERBMT). The proposed methodology is flexible and scalable. The main advantages of HERBMT are that it combines the advantages of RBMT and EBMT, and it can be applied to other languages with minor modifications. EBMT extracts an example of target language sentences that are analogous to input source language sentences. The extraction of appropriate translated sentences is preceded by an analysis stage for the decomposition of input sentences into appropriate fragments. RBMT is used when examples of the source language to be translated into the target language are not found in the machine database. In addition, RBMT can be applied to other languages with some modifications. The database has been designed to be flexible where most of the rules are defined in the database tables in order to generalize the code. The OAK Parser is used to analyze the input English text to get the part of speech (POS) for each word in the text as a pre-translation process. Validation rules have been applied in both the database design and the programming code in order to ensure the integrity of data. A major design goal of this system is that it will be used as a stand-alone tool, and can be integrated with a general machine translation system for English sentences. The evaluation is carried out on 250 independent test suites, and the analysis indicates that HERBMT achieved good performance with an average of 97.2% precision.,Ph.D
Pages:	333
Call Number:	QA76.9.N38.A5647 2013 3
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.