2014 TRB Data Competition
Sponsoring Committee:
ABJ80 Statistical Methods
Co-Sponsoring Committees:
ABJ00 Data Section
ABJ70 Artificial Intelligence and Advanced Computing Applications
ABJ20 Statewide Transportation Data and Information Systems
ABJ50 Information Systems and Technology
AND10 Vehicle User Characteristics
AND30 Simulation and Measurement of Operator Performance
The Transportation Research Board is pleased to announce the 2014 Transportation Data Competition workshop, to be held at the 93rd TRB Annual meeting. The scope of the competition is to evaluate the appropriateness of various data analytic methodologies – be they statistics, artificial intelligence, or advanced data visualization– for forecasting trends in safety, with particular attention to missing and incomplete data. The objective is to identify modelling tools in statistics and artificial intelligence for forecasting and to disseminate knowledge on “best practices” in transportation data modeling. The dataset used for this competition will be based on a data set on driver behavior, using data from the National Advanced Driving Simulator (NADS), in which drivers traverse an intersection with a traffic signal that may change from green to yellow phase.
For complete details visit here
Data contest guideline and Data description (updated on Oct 13, 2013): Download
Data variable definition (updated on Nov 18, 2013): Download
Questions & Answers (updated on Nov 18, 2013): Download
Data set (updated on Nov 19, 2013): Download
Results
The competition received 31 unique submissions from individuals in industry and academia, from professors, students, and researchers.
There were some really outstanding work, and we were highly impressed by all the analytical tools that were used. Based on the many diverse backgrounds, we have decided to give the following six awards based on their high rankings for model quality, scientific rigor, paper quality, and overall presentation of results.
Best Overall Paper: Jeff Keller, RSG
Best Professor-Led Paper: Naveen Eluru and Shamsunnahar Yasmin, McGill U.
Best Researcher-Led Paper: Huimin Xiong, Prabha Narayanaswamy, Shan Bao, and Carol Flannagan, UMTRI
Best Student-Led Paper: Behzad Karimi, UIC
Best Industry Paper: Nikhil Sikka and Tristan Cherry, RSG
Industry Honorable Mention: Micahel Abromovich, Swapnil, Rajiwade, and Dwayne Henclewood, BAH
ABJ80 Statistical Methods
Co-Sponsoring Committees:
ABJ00 Data Section
ABJ70 Artificial Intelligence and Advanced Computing Applications
ABJ20 Statewide Transportation Data and Information Systems
ABJ50 Information Systems and Technology
AND10 Vehicle User Characteristics
AND30 Simulation and Measurement of Operator Performance
The Transportation Research Board is pleased to announce the 2014 Transportation Data Competition workshop, to be held at the 93rd TRB Annual meeting. The scope of the competition is to evaluate the appropriateness of various data analytic methodologies – be they statistics, artificial intelligence, or advanced data visualization– for forecasting trends in safety, with particular attention to missing and incomplete data. The objective is to identify modelling tools in statistics and artificial intelligence for forecasting and to disseminate knowledge on “best practices” in transportation data modeling. The dataset used for this competition will be based on a data set on driver behavior, using data from the National Advanced Driving Simulator (NADS), in which drivers traverse an intersection with a traffic signal that may change from green to yellow phase.
For complete details visit here
Data contest guideline and Data description (updated on Oct 13, 2013): Download
Data variable definition (updated on Nov 18, 2013): Download
Questions & Answers (updated on Nov 18, 2013): Download
Data set (updated on Nov 19, 2013): Download
Results
The competition received 31 unique submissions from individuals in industry and academia, from professors, students, and researchers.
There were some really outstanding work, and we were highly impressed by all the analytical tools that were used. Based on the many diverse backgrounds, we have decided to give the following six awards based on their high rankings for model quality, scientific rigor, paper quality, and overall presentation of results.
Best Overall Paper: Jeff Keller, RSG
Best Professor-Led Paper: Naveen Eluru and Shamsunnahar Yasmin, McGill U.
Best Researcher-Led Paper: Huimin Xiong, Prabha Narayanaswamy, Shan Bao, and Carol Flannagan, UMTRI
Best Student-Led Paper: Behzad Karimi, UIC
Best Industry Paper: Nikhil Sikka and Tristan Cherry, RSG
Industry Honorable Mention: Micahel Abromovich, Swapnil, Rajiwade, and Dwayne Henclewood, BAH