Road traffic accident reports have important information for analyzing and preventing road traffic accidents. Early information extraction methods of road traffic accident reports were mainly manual reading and manual input, which cost a lot of manpower and material resources and isn’t conducive to the aggregation and mining of accident information. Therefore, this paper innovatively proposes a key information extraction method of road traffic accident via integration of rules and SkipGram-BERT. Firstly, the paper constructs the key information index system of road traffic accidents based on the characteristics of road traffic accident reports. At the same time, the unstructured road traffic accident reports are processed by sentence segmentation and word segmentation. Secondly, for indicators with stronger regularity, the paper builds extraction rules for each indicator. For indicators with more diverse expressions, the paper proposes the SkipGram-BERT information extraction model. Finally, this paper applies the key information extraction method of road traffic accidents to practical cases to verify the effectiveness of the model. The result shows that the model proposed in this paper has an average accuracy of 85.52% for information extraction, which can more accurately extract key road traffic information. This method can help road traffic managers to quickly obtain the key information of road traffic accidents, and greatly improve the efficiency of road traffic accident industry management.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Key Information Extraction Method Study for Road Traffic Accidents via Integration of Rules and SkipGram-BERT


    Weitere Titelangaben:

    Lecture Notes on Data Engineering and Communications Technologies


    Beteiligte:
    Hu, Zhengbing (Herausgeber:in) / Wang, Yong (Herausgeber:in) / He, Matthew (Herausgeber:in) / Li, Cuicui (Autor:in) / Zhang, Jixiu (Autor:in) / Li, Baidan (Autor:in) / Xu, Zhiyuan (Autor:in)

    Kongress:

    The International Symposium on Computer Science, Digital Economy and Intelligent Systems ; 2022 ; Wuhan, China November 11, 2022 - November 13, 2022



    Erscheinungsdatum :

    2023-01-29


    Format / Umfang :

    15 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Defending road traffic accidents

    IBC UK Conferences Limited | British Library Conference Proceedings | 2000


    Roads, traffic, accidents and road research

    Rigden, P.J. | Engineering Index Backfile | 1964


    Road traffic accidents in a Swedish municipality

    Schelp, L. / Ekman, R. | Elsevier | 1990


    Disastrous but preventable: Road traffic accidents

    Davidson, Patricia M. / Dharmaratne, Samath D. | Taylor & Francis Verlag | 2016


    Road-user behavior and traffic accidents

    Näätänen, Risto ;Summala, Heikki | SLUB | 1976