Scholarlink

In metropolitan cities, road safety is of utmost importance, especially in densely populated places like Kolkata, India, where the frequency and severity of traffic accidents present significant challenges. Numerous accidents happen every day that should be decreased. From the crash data, hidden patterns are extracted using the association rule mining (ARM) approach. The use of the method in the transportation sector is examined in this study, with an emphasis on finding hidden patterns, correlations, and linkages to improve decision-making in resource allocation, road safety, traffic management, and route optimization. We use extensive datasets from 2017–2020 and 2021–2023. To find the frequently occurring itemsets in the dataset, we next utilize the Predictive Apriori technique. After that, regression analysis and advanced cluster analysis are employed. High R2 values have shown that models like Random Forest and XGBoost have excellent prediction ability. The study findings not only provide insight into the trends and predictors of traffic accidents, but they also play a vital role in the creation of focused, data-driven actions meant to improve road safety in Kolkata. This study uses in-depth spatiotemporal analysis to offer insightful information that can assist strategic planners and policymakers in reducing traffic accidents in metropolitan areas.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Unveiling Hidden Patterns by Integrating Association Rule Mining and Regression Models for Intelligent Transportation Systems

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Moumita Mishra,
  • Soumya K Ghosh,
  • Bhargab Maitra

摘要

In metropolitan cities, road safety is of utmost importance, especially in densely populated places like Kolkata, India, where the frequency and severity of traffic accidents present significant challenges. Numerous accidents happen every day that should be decreased. From the crash data, hidden patterns are extracted using the association rule mining (ARM) approach. The use of the method in the transportation sector is examined in this study, with an emphasis on finding hidden patterns, correlations, and linkages to improve decision-making in resource allocation, road safety, traffic management, and route optimization. We use extensive datasets from 2017–2020 and 2021–2023. To find the frequently occurring itemsets in the dataset, we next utilize the Predictive Apriori technique. After that, regression analysis and advanced cluster analysis are employed. High R2 values have shown that models like Random Forest and XGBoost have excellent prediction ability. The study findings not only provide insight into the trends and predictors of traffic accidents, but they also play a vital role in the creation of focused, data-driven actions meant to improve road safety in Kolkata. This study uses in-depth spatiotemporal analysis to offer insightful information that can assist strategic planners and policymakers in reducing traffic accidents in metropolitan areas.