Content:An important task of natural language processing is part of speech tagging. We have a lot of research on Arabic, but due to the complexity of these tasks and the characteristics of Arabic, these performances have not reached the standard level. Any part-of-speech tagger is designed to assign a word class to each word in the text, or to match a word in the text, for example, in an Arabiclanguagea wordcanbea verb, noun, or particle. Therefore, the Arabic POS-Tagger should match each word in the text with a word indicating its category (verb, noun, or particle). Labels added to any text are filled with several purposes and used to determine language and analytical word frequency, syntactic structure, and other analysis.
Mohamed Labidi proposed a new combination of improved Arabic POS tags, which was published in the journal Autonomous Intelligence. In this work, Mohamed Labidi studied the combination of two different methods of Arab POS-Taggers. The first is based on the maximum value of entropy; the second is based on statistics/rules. In addition, Mohamed Labidi has added a knowledge-based approach to annotating Arabic particles. Mohamed Labidi's vision increased the accuracy of the marker.
Click here to read more: