A Pilot Study to Integrate Mobility Data Collection APPs with Personalized Recommendation Systems

Details

• Creators:
  Zhang, Congjing ; Huang, Shuai
• Corporate Creators:
  Center for Understanding Future of Travel Behavior and Demand (TBD) National University Transportation Center (UTC) ; University of Washington. Dept. of Industrial and Systems Engineering
• Corporate Contributors:
  United States. Department of Transportation. University Transportation Centers (UTC) Program ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology
• Subject/TRT Terms:
  Data Collection Mobile Applications Optimization Recommendations Sensors Smartphones Travel Behavior
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Edition:
  Final Report, 2023-2025
• Corporate Publisher:
  Center for Understanding Future of Travel Behavior and Demand (TBD) National University Transportation Center (UTC)
• Abstract:
  This pilot study explores the integration of mobility data collection applications with personalized recommendation systems to enhance user-specific travel suggestions and promote sustainable transportation behavior. We leverage the open-source NREL OpenPATH platform to collect detailed mobility data including travel trajectories, inferred modes, and user annotations from a cohort of participants over a one-year period. Our analysis identifies several challenges in the sensed data, including misclassification of transportation modes and inaccurate trip segmentation, which highlight the need for improved sensing and user interface design. To establish a foundation for recommendation modeling, we evaluate the ItemKNN algorithm on a filtered subset of the Yelp dataset. Results indicate that user history richness improves performance across precision, recall, and nDCG metrics. Building on these insights, we develop a prototype personalized recommendation system that dynamically suggests Points-of-Interest (POIs) based on location, time, and user behavior. This system incorporates a reward mechanism that offers incentives to encourage the adoption of suggested alternatives and collects user feedback to iteratively refine future recommendations. The findings suggest that integrating personalized recommendations with mobility tracking can create a closed-loop system where improved data quality enhances recommendations, which in turn drives user engagement and behavior change. This pilot demonstrates the feasibility of such integration and offers a blueprint for future research. Applications of this framework include scalable interventions for promoting active travel, reducing carbon emissions, and supporting individualized transportation planning through intelligent, adaptive systems.
• Format:
  PDF
• Funding:
  69A3552344815, 69A3552348320
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:259cddc794c473eaf58c00dbd500ae2aeb79b20b44980e3bfb1e23e7d2eb9da71996be7f9c2e345fbfbd0d6662d1c7148c7a57cce9f1b648576190d693fed5e3
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/87070/dot_87070_DS1.pdf
• File Type:
  [PDF - 1.62 MB ]