A Named Entity Recognition Model Based on Entity Trigger Reinforcement Learning

Abstract

Named entity recognition is a practical approach to automatically identifying named entities in text and data. Towards the vast amount of data generated in our daily life, Artificial Intelligence (AI) with economical but powerful computing resources are inevitably becoming the most appropriate method for name entities classification. However, the results of currently popular methods may also lack the aiming super high accuracy to specific data and the interests of the subscribers. This paper proposes a named entity recognition model based on entity trigger reinforcement learning for automatic Chinese recognition. Unlike existing named entity recognition methods, the proposed method can support multiple inputs. The accuracy proof and performance evaluation show that the proposed method is provable robotic in entity categories classification and efficient in practice.