Diacritics of arabic natural language processing (ANLP) and its quality assessment

Alzand, Abdulrahman Ahmed and Ibrahim, Rosziati (2015) Diacritics of arabic natural language processing (ANLP) and its quality assessment. In: Proceedings of the 2015 International Conference on Industrial Engineering and Operations Management (IEOM 2015), 3-5 March 2015, Dubai, United Arab Emirates (UAE).

[img]PDF
727Kb

Abstract

Arabic language is a unique language because of its pronunciation of the word written. The formation of more than two Arabic letters will translate to a different meaning. This paper presents an alternative translator from Arabic word to English word by using an equation for the formation of the Arabic letters. The morphological model of Arabic language is used in order to use the equation to put diacritics according to Arabic grammatical rules. A system is then developed based on the morphological model. The input of the system is the Arabic word. The system uses morphological Arabic Natural Language Processing (ANLP) and translates the Arabic word into English word. The output of the system shows percentage of the successful translated words. We tested our system by using 11 Arabic words. The results show that 9 out of 11 Arabic words (81.81%) were successfully translated by conducting the process of diacritics in the morphological model of Arabic language. We also perform a quality assessment to calculate the efficiency and effectiveness metrics based on the ANLP developed. Based on the result, the system can be used as an alternative translator from Arabic word to English word

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:natural language processing (NLP); morphological model; efficiency and effectiveness metrics;
Subjects:Q Science > QA Mathematics > QA76 Computer software
Divisions:Faculty of Computer Science and Information Technology > Department of Software Engineering
ID Code:6754
Deposited By:Mrs. Nurhayati Ali
Deposited On:13 Apr 2015 16:38
Last Modified:13 Apr 2015 16:38

Repository Staff Only: item control page