With the gradual advancement of human exploration of space, textual information related to spacecraft has grown exponentially. It's necessary to achieve the automatic acquisition of relevant information, explore the information extraction of the spacecraft domain knowledge map, and solve the problem lacking existing labeled data sets. The book “Encyclopedia of World Spacecraft” is selected as the data source to build a spacecraft-related data set and the entities in the field are divided into spacecraft, research countries, research institutions and platforms by the spacecraft text and experts. Based on the TFT model composed of text representation, feature extraction and sequence labeling, set up multiple sets of comparative experiments to find a model with the best recognition effect. Comparative experiments show that the Bert-BiGRU-CRF model combined with transfer learning can fully learn the characteristic content of the spacecraft field and has a better recognition effect than other models in terms of accuracy, recall and F1 value.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Research on Named Entity Recognition Based on Bert-BiGRU-CRF model in Spacecraft Field


    Contributors:
    Tong, Boan (author) / Pan, Jian (author) / Zheng, Leixin (author) / Wang, Liyang (author)


    Publication date :

    2021-09-24


    Size :

    350071 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    aeroBERT-NER: Named-Entity Recognition for Aerospace Requirements Engineering Using BERT

    Ray, Archana Tikayat / Pinon-Fischer, Olivia J. / Mavris, Dimitri N. et al. | TIBKAT | 2023


    aeroBERT-NER: Named-Entity Recognition for Aerospace Requirements Engineering using BERT

    Tikayat Ray, Archana / Pinon-Fischer, Olivia J. / Mavris, Dimitri N. et al. | AIAA | 2023


    BERT-BiLSTM-BiGRU-CRF: Ensemble Multi Models Learning for Product Review Sentiment Analysis

    Mouthami, K. / Anandamurugan, S. / Ayyasamy, S. | IEEE | 2022


    IDCNN-CRF-based domain named entity recognition method

    Yu, Bihui / Wei, Jingxuan | IEEE | 2020


    The Span-Based Transformer Classifier for Named Entity Recognition

    Li, Yangfan / Sun, Xin / Li, Hongao et al. | Springer Verlag | 2022