Traffic violations are a major cause of traffic accidents, yet current research falls short in comprehensively analysing these violations and the  named entity method fails to extract the name of traffic violation events from records, thereby lacking in providing guidance for managing urban traffic violations. By expanding the People’s Daily dataset from 71,456 words to 95,291 words, the BERT-CRF (Bidirectional Encoder Representations from Transformers-Conditional Random Field) model achieves an accuracy rate of 88.53%, a recall rate of 92.90% and an F1 score of 90.66%, successfully identifying event, time and location named entities within traffic violations. The data of traffic violations is then enhanced through forward geocoding and the Bayesian formula, and traffic violations are analysed from time, space, administrative region, gender and weather, to provide support for the dynamic allocation of law enforcement forces on traffic scenes and the precise management of traffic violations.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Analysis of Beijing Traffic Violations Based on the BERT-CRF Model


    Beteiligte:
    Jie Li (Autor:in) / Yuntao Shi (Autor:in) / Shuqin Li (Autor:in)


    Erscheinungsdatum :

    2024




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Unbekannt




    Construction of Knowledge Graph Based on Traffic Violations in Beijing

    Li, Jie / Shi, Yuntao / Li, Shuqin et al. | IEEE | 2022


    Traffic violations analysis: Identifying risky areas and common violations

    Ben Laoula, El Mehdi / Elfahim, Omar / El Midaoui, Marouane et al. | Elsevier | 2023

    Freier Zugriff

    Detection of traffic violations

    SCHENKEN CHRISTOPHER T / DAVIDSON MARK J | Europäisches Patentamt | 2015

    Freier Zugriff

    Detection of traffic violations

    SCHENKEN CHRISTOPHER T / DAVIDSON MARK J | Europäisches Patentamt | 2017

    Freier Zugriff

    Traffic accidents and violations

    Engineering Index Backfile | 1956