Research on Named Entity Recognition Method Based on Option Hierarchical Reinforcement Learning

The educational knowledge graph plays an important role in smart education, and how to effectively recognize named entities is the first important part of constructing educational knowledge graph. This study proposes a named entity recognition model based on Option hierarchical reinforcement learning to realize named entity recognition for domain-specific knowledge in higher education. The innovations of this study are reflected in the following two aspects: First, based on the domain knowledge characteristics, a multi-head attention mechanism based on weighted combinatorial similarity is designed to realize further screening of information. Second, the concept of hierarchical reinforcement learning is introduced into the naming recognition model, and the Option-Critic framework is improved and applied to named entity recognition to improve the accuracy of naming recognition. In order to verify that the model proposed in this study is valid, Computer Networks, a core foundation course for university computer science majors, is used as an example. Comparison experiments are conducted using the traditional Bert-BiLSTM-CRF model and the present model, and the experiments prove that the model proposed in this study is feasible and effective.