Volume : 4, Issue : 3, March - 2015

Computational Method for Accurate Classification of Arabic Texts Based on Arabic Phonetic Transcription

Dr. Ahmad Abdulqader Abuseeni

Abstract :

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