Machine Learning Tools for Informing Transportation Technology and Policy [Research Brief]

Details

• Creators:
  Cummings, Mary
• Corporate Creators:
  Duke University
• Corporate Contributors:
  Collaborative Sciences Center for Road Safety ; United States. Department of Transportation. Federal Highway Administration
• Subject/TRT Terms:
  • Alternatives analysis
  • Data analysis
  • Evaluation and assessment
  • Fatalities
  • Logistic regression analysis
  • Machine learning
  • Neural networks Explainable artificial intelligence
  • Feature selection
• Publication/ Report Number:
  CSCRS-R10
• Resource Type:
  Brief
• Geographical Coverage:
  United States
• Corporate Publisher:
  Collaborative Sciences Center for Road Safety
• Abstract:
  Little work has been conducted to study practitioner induced subjectivities introduced into machine learning applications, which is important in understanding causes, influences, and methods for avoidance, particularly in transportation settings. To help fill this gap, this study uses two transportation datasets to examine car and pedestrian crash fatalities, deploying two different machine learning techniques of low and high complexity (logistic regression and neural networks).
• Format:
  PDF
• Funding:
  69A3551747113
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/78264/dot_78264_DS1.pdf
• File Type:
  [PDF - 392.52 KB]
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:cc9ff507c34a0ed25192bebc66440812f9c6c28709a7c34ccace2c935865cf9757c88be3d89468c746da0654691740dd751a6608cf13e5ce11f34bb45b11f840