A Model of Chinese Named Entity Recognition in the Field of Oil and Gas Exploration Using Domain Dictionary
Bingxu Liu, Zhiguang Wang, Yize Ding, Qiang Lü · 2021
In view of the scarcity of labeled data in the task of Chinese named entity recognition in the field of oil and gas exploration, the large number of entity types and the difficulty of pre-defining, and the problem of entity combination and nesting in many domain entities, this paper proposes a model of Chinese named entity recognition in the field of oil and gas exploration using domain dictionary (OGDNER) which uses the domain dictionary to match domain documents, and designs a boundary correction algorithm to reduce noises generated in the dictionary matching process. In addition, the “Break-Tie” labeling mode is proposed as well as the input of the neural network is optimized to improve the effect of model. Finally, the effectiveness of the model is verified by comparing it with other models on the data set of oil and gas exploration.