The paper provides a comprehensive survey of deep learning-based named entity recognition (NER) methods. It begins by introducing the problem and the limitations of traditional NER methods, which heavily rely on manual feature engineering and struggle with large, complex datasets. The paper then categorizes commonly used NER datasets into three classes based on the complexity of named entities and reviews typical deep learning-based NER methods according to the development history of deep learning models. It conducts an in-depth analysis and comparison of methods that have achieved outstanding performance on representative datasets, reproduces and analyzes the recognition results of some typical models on three different types of datasets, and concludes by offering insights into future trends in NER development. The paper aims to provide a visual basic cognition of the NER process, reduce the complexity of NER research, and enhance the feasibility of NER research through experimental comparisons.The paper provides a comprehensive survey of deep learning-based named entity recognition (NER) methods. It begins by introducing the problem and the limitations of traditional NER methods, which heavily rely on manual feature engineering and struggle with large, complex datasets. The paper then categorizes commonly used NER datasets into three classes based on the complexity of named entities and reviews typical deep learning-based NER methods according to the development history of deep learning models. It conducts an in-depth analysis and comparison of methods that have achieved outstanding performance on representative datasets, reproduces and analyzes the recognition results of some typical models on three different types of datasets, and concludes by offering insights into future trends in NER development. The paper aims to provide a visual basic cognition of the NER process, reduce the complexity of NER research, and enhance the feasibility of NER research through experimental comparisons.