Document Type
Technical Report
Publication Date
11-13-2024
Keywords
Travel behavior, passenger driver classification, mobile biometrics
Digital Object Identifier (DOI)
https://doi.org/10.5038/CUTR-NICR-Y3-3-10
Abstract
Our study investigates smartphone sensor data for classifying drivers and passengers, aligning with our original proposal to enhance multimodal transportation data collection via the OneBusAway app. We address challenges in balancing data quality with smartphone resource constraints and find that shorter data collection windows significantly improve classification performance. Additionally, feature standardization and selection enhance accuracy, particularly during extended data periods, supporting our goal of refining data quality from open-source platforms. These insights advance our objectives by optimizing data efficiency and supporting the real-world applicability of intelligent models in transportation safety and efficiency.
Scholar Commons Citation
Lozano, Wilson and Neal, Tempestt, "Efficient Smartphone Sensor Analysis for Behavioral Profiling in Transportation Research: A Case Study of Driver and Passenger Classification" (2024). Research Reports. 51.
https://digitalcommons.usf.edu/cutr_nicr/51
Policy Brief
