Computational Method for Accurate Classification of Arabic Texts Based on Arabic Phonetic Transcription
Saudi Arabia · 2015
Arabic auto translation is challenging not only because of the inter-language transfer errors that may result, but also because of the lack the absence of an effective, accurate and sufficient retrieval system. In the Arabic language, many words are written in the same letters but have different phonetic transcriptions and meanings. As such, Arabic has a very complicated morphological structure with prefixes, infixes, and suffixes. In this focus of this scientific paper primarily revolves on the analysis of Arabic phonetic transcription based on a Bayesian Method support structure as well as Support Vector machine algorithms to classify Arabic texts. In this analytical paper, analysis focuses on systems than are supported by word pronunciation and diacritical marks. In addition to vowel letters, ablaut and slurring voice units, this paper takes into consideration the diacritical marks which make classification of Arabic texts based on Arabic phonetic transcription more efficient. To this end, experiemtns in this research highlight promising outcomes.