First Order Hidden Markov Model for Automatic Arabic Name Entity Recognition

Abstract

Name Entity Recognition (NER) is an important process used for several type of applications such as Information Extraction, Information Retrieval, Question Answering, text clustering, etc. It is intended to identify and classify name entities from a given text. NER is performed by using a rule-based approach that relies on human intuitive or machine learning methods such as Hidden Markov Model (HMM), Maximum Entropy (ME), and Decision tree (DT). In this paper, we describe a model based on the first order HMM to recognize name entity in the Arabic language. The model is based on stemming process that solves Arabic's inflection problem and ambiguity. To the best of our knowledge, no work uses this approach for the Arabic language has been reported. General Terms Natural Language Processing (NLP), Arabic language.