استخدام اللغة العربية في نظم استرجاع المعلومات

Abstract

Dissertation Title: The use of Arabic Language in Information Retrieval Systems Prepared by: Nadia Mustafa Elidrous Ahmed Supervisor: Professor Radia Adam Mohammed University: University of Khartoum - Faculty of Arts- Department of Library and Information Sciences Pages: 309 p Year: 2007 The aim of this research is to study the characteristics of Arabic language which affect the information retrieval process and to identify the problems arising from the use of Arabic language in the different information systems then come up with recommendations to solve such problems. Also to compare the tools and search strategies employed in the information retrieval systems as regards to their influence on the precision ratio of the retrieval documents and to study the implications of the application or none application of Arabic language processing technologies in computers (spelling and grammar checking, and morphological analysis … etc) on the processing of Arabic language characteristics (synonymy, homonym, antonymy … etc). The research adopted the descriptive and experimental methodologies by executing a series of investigations on a selected sample of the Arabic and Arabic language supporting retrieval systems ( integrated, none integrated, open source and the search engine available in the internet ) using different search implement levels (word, prefix and sub fix and the word, two words, a sentence and the root of the word ) and employing different search methods ( the natural language, Boolean, search in specified fields, search using quotation signs ). All this to measure the effect of using Arabic language characteristics (synonymy, and homonym … etc) in addition to the spelling mistakes and miswriting of Arabic words and in regards to the comparison between the precision ratio (finding the percentage of the records concerning the search topic to the information retrieval system tacking in consideration the processing of Arabic language technologies of the characteristics of the language, the Arabic definition article ( أل ) and the different levels of research mentioned above. The study depended on a sample selected from the following information retrieval systems: the UNESCO mechanized documentation system. (CDSISIS), the Horizon system, the Greenstone system, the Arabic search engine (Alidrisi and Ayna) and the Arabic language supporting search engine (Google). The study report comprises of five chapters. The first chapter is about the Arabic language, its characteristics, properties, systems and sciences. The second chapter handles the use of Arabic language in computers, arabicization of computers, standardization of Arabic language in computers, the mechanized processing of Arabic language in computers, the techniques of processing the Arabic language in computers and the problems of information retrieval in the Arabic language. The third chapter concentrates on the commercial and none commercial information retrieval systems employed in the information institutions with stress on the services provided, language used, the technologies affecting the systems and the steps followed and tools used in searches. Sample of these systems are mentioned. Chapter four deals with the Arabic information retrieval systems and the Arabic information retrieval and Arabic language supporting systems available in the internet. The advantages and disadvantages, the services of the internet and the technologies influencing it. The chapter also concentrates on the search tools, the search engine, the directories and portals of search, examples of these are also mentioned. Chapter five contains an analytical study aimed to investigate the effect of using the characteristics of Arabic language in the information retrieval systems. The study culminated in many results. The most important of which are: Despite the increase of the Arabic content in the internet, the proper and accurate dealing with this content and making use of it as much are possible is still limited. This is due to inefficiency of the search mechanism of the Arabic language which takes in consideration all the characteristics of the Arabic language. Many conclusions are made e.g: the need for a compulsory standard for Arabic encoding characters when publishing information in the internet.