The role of the morphological component in building the electronic dictionary of the Arabic language - Sources database-

Abstract

This research will focus on the role of morphological weight and balance in building the electronic dictionary of the Arabic language, and how the neglect of this component or its poor application led to the weakness of many Arabic programs, especially in building the morphological analyzer and generator, despite the great efforts spent in computerizing the Arabic language. . For this reason, we do not find today Arabic computer programs capable of accomplishing the task of machine translation to a reasonable, or close to reasonable, rate, nor do we find programs for visual reading of Arabic characters that comfortably fulfill the purpose, and there are many other programs, and this - in our view - is only to reduce this problem. Software plays the role of the morphological component of the language. It will also focus on the fusion nature of the Arabic language, unlike European languages, which are described as agglutinative languages, and will monitor the regularity of this component in word formation in generation and analysis, and its ability to distribute morphemes within the structure of Arabic roots to make them lexical entries that become susceptible to semantic classification, syntactic composition, and pragmatic diversity. . We will also provide a detailed definition of the concept of the electronic dictionary and its applications in natural languages, focusing on the morphological (algorithmic) characteristics of the Arabic language from the point of view of linguistic engineering, and highlighting the importance of this component and its necessity for building the automated syntactic analyzer for the Arabic language, as it has become clear to us that it is difficult to advance in building the rest of the analyzers. The mechanism of the Arabic language before completing the description of the morphological component in a linguistic and geometric manner. We will propose a computer-linguistic model that we apply to a database of Arabic sources (250,000 encrypted morphological entries), to address this situation based on the theoretical approach that makes the morphological weight the deep structure of the Arabic word, while making the morphological balance its superficial structure. We will conclude the research by analyzing the statistics of all morphological entries generated morphologically in the Arabic language, which include about 2,500,000 lexical entries generated from the total number of Arabic roots actually used (12,000 roots), with about 800 morphological forms, distributed over the following parts of the word: subject nouns, verbs, infinitives, Derivatives.