scholarly journals Automatic Translation of Arabic Sign to Arabic Text (ATASAT) System

Author(s):  
Abdelmoty M.Ahmed ◽  
Reda Abo Alez ◽  
Muhammad Taha ◽  
Gamal Tharwat
1977 ◽  
Vol 16 (03) ◽  
pp. 144-153 ◽  
Author(s):  
E. Vaccari ◽  
W. Delaney ◽  
A. Chiesa

A software system for the automatic free-text analysis and retrieval of radiological reports is presented. Such software involves: (1) automatic translation of the specific natural language in a formalized metalanguage in order to transform the radiological report in a »normalized report« analyzable by computer; (2) content processing of the normalized report to select desired information. The approach used to accomplish point (1) is described in detail referring to a specific application.


2010 ◽  
Vol 12 (1-2) ◽  
pp. 337-314
Author(s):  
ʿAbd Allāh Muḥammad al-Shāmī

The question of clarifying the meaning of a given Arabic text is a subtle one, especially as high literature texts can often be read in more than one way. Arabic is rich in figurative language and this can lead to variety in meaning, sometimes in ways that either adhere closely or diverge far from the ‘original’ meaning. In order to understand a fine literary text in Arabic, one must have a comprehensive understanding of the issue of taʾwīl, and the concept that multiplicity of meaning does not necessarily lead to contradiction. This article surveys the opinions of various literary critics and scholars of balāgha on this issue with a brief discussion of the concepts of tafsīr and sharḥ, which sometimes overlap with taʾwīl.


2004 ◽  
Vol 6 (2) ◽  
pp. 170-183
Author(s):  
Hassan al-Shafīe

The present study discusses the cultural and intellectual movement, now on the point of prevalence in the contemporary Islamic world, which adopts the Western ‘hermeneutical method’ and applies it to the Qur'an in particular, and Islamic religious texts in general. The author shows this movement's complete disregard for the established principles of tafsīr, the traditional Arab-Islamic rules of Qur'anic interpretation and the related Prophetic aḥādīth as preserved in the authenticated Sunna. The author argues that the ‘hermeneutical method’ starts from the preconceived notion that the Islamic heritage is male-centred and biased against women, both theoretically and practically, and, on this basis, proposes that the time has come for an intellectual break with this premise and the re-interpretation of the Qur'an and faith in the light of Western Christian hermeneutics. This paper proposes that this method fails to take historical events and the civilisational Islamic experience into account.


Author(s):  
Aliona Kolesnichenko ◽  
Natalya Zhmayeva

The article is devoted to the analysis of grammatical difficulties encountered in the process of automatic translation. The paper discusses the advantages and disadvantages of the SDL Trados automatic translation service. The types of grammatical errors when translating scientific and technical texts in SDL Trados are classified, the ways of overcoming them are outlined. Key words: scientific and technical literature, automatic translation, grammatical difficulties.


2020 ◽  
Vol 17 (3) ◽  
pp. 299-305 ◽  
Author(s):  
Riaz Ahmad ◽  
Saeeda Naz ◽  
Muhammad Afzal ◽  
Sheikh Rashid ◽  
Marcus Liwicki ◽  
...  

This paper presents a deep learning benchmark on a complex dataset known as KFUPM Handwritten Arabic TexT (KHATT). The KHATT data-set consists of complex patterns of handwritten Arabic text-lines. This paper contributes mainly in three aspects i.e., (1) pre-processing, (2) deep learning based approach, and (3) data-augmentation. The pre-processing step includes pruning of white extra spaces plus de-skewing the skewed text-lines. We deploy a deep learning approach based on Multi-Dimensional Long Short-Term Memory (MDLSTM) networks and Connectionist Temporal Classification (CTC). The MDLSTM has the advantage of scanning the Arabic text-lines in all directions (horizontal and vertical) to cover dots, diacritics, strokes and fine inflammation. The data-augmentation with a deep learning approach proves to achieve better and promising improvement in results by gaining 80.02% Character Recognition (CR) over 75.08% as baseline.


Sign in / Sign up

Export Citation Format

Share Document