Search Results for "tashkeela"

A version of the Tashkeela Arabic diacritized text dataset cleaned from the non-Arabic content and the undiacritized text, then divided into training, development, and testing sets. The cleaning process includes removing the XML tags and strange symbols, as well as fixing diacritics errors. After that, the tokenization is performed while focusing on the extraction of the Arabic words. The result is a space-separated tokens file, where the words and the numbers are separated

In this corpus: 10 essays containing 752 sentences (with a total of 4,160 words). The essays were selected from different collections of partially or totally diacritic Arabic texts, all of which are available in the Tashkeela corpus. Texts in this corpus have been used in the evaluation of AGD checker. There are two types of texts in this corpus: 1- Texts without errors to evaluate AGD in terms of detecting and correcting errors that we do not know about before the checking process 2...

Tashkeela: Arabic discritization Corpus, Resource, Arabic vocalized texts: نصوص عربية مشكولة =========== Contains Arabic text vocalized . Text -format; 75.6 millions words Please cite this resource as: T. Zerrouki, A. Balla, Tashkeela: Novel corpus of Arabic vocalized texts, data for auto-diacritization systems, Data in Brief (2017), http://dx.doi.org/10.1016/j.dib.2017.01.011 Data in Brief ∎ ( ∎∎∎∎ ) ∎∎∎ – ∎∎∎