Survey For Arabic Part of Speech Tagging based on Machine Learning

Abstract

      The Arabic Language is the native tongue of more than 400 million people around the world,  it is also a language that carries an important religious and international weight.  The Arabic language has taken its share of the huge technological explosion that has swept the world, and therefore it needs to be addressed with natural language processing applications and tasks.

This paper aims to survey and gather the most recent research related to Arabic Part of Speech (APoS), pointing to tagger methods used for the Arabic language, which ought to aim to constructing corpus for Arabic tongue. Many AI investigators and researchers have worked and performed POS utilizing various machine-learning methods, such as Hidden-Markov-Model (HMM), Brill, Maximum-Match (MM), decision tree, bee colony, Neural-Network (NN), and other hybrid methods.

This survey groups a number of published papers based on the Arabic Language Applications (ALP) towards tagging related problems utilized and approaches with the difference between types of tags used. It addresses and tries to identify the gaps in the current studies putting a foundation for future studies in this field.