Webology (ISSN: 1735-188X) Volume 18, Number 4, 2021 250 http://www.webology.org A Survey On Deep Learning Approaches For Named Entity Recognition Hamid Sadeq Mahdi Alsultani 1,2 , Ahmed H. Aliwy 2 1 Department of Computer Science, College of Basic Education, University of Diyala, Diyala, Iraq. 2 Department of Computer Science, Faculty of Computer Science and Mathematics, University of Kufa, Najaf, Iraq. __________________________________________________________________ Abstract Named Entity Recognition (NER) is considered as a task of Information Extraction (IE) which is effective for improving the efficiency of a variety of Natural Language Processing (NLP) tasks, including Relation Extraction (RE), Question Answering (QA), Information Retrieval (IR), etc.NER tries to identify and classify named entities from a specified text, like persons, locations, and organizations, etc. Many researchers have discussed this problem through a variety of approaches, including rule-based and machine learning-based approaches. In recent years, many NER approaches that are based on deep learning have been proposed and improved to obtain precise results .In this paper, a survey of deep learning approaches for NER was presented. Also, the datasets and the evaluation metrics used in each approach were demonstrated. Then, a discussion was provided about the surveyed articles in terms of deep learning NER approaches together with the datasets and evaluation metrics used with each approach. Finally, a conclusion was presented. Keywords__ named entity recognition, information extraction, natural language processing, deep learning 1 INTRODUCTION In 1996, Named entity recognition (NER) was first presented as a task of information extraction at the Message Understanding Conferences, and it was considered as a significant Natural Language Processing (NLP) task by academics [1].NER can be defined as the task of extracting and classifying named entities into pre-defined types or classes such as persons, locations, organizations, dates, diseases, weapons, genes, drugs, etc. in specific domain's texts [2].NER