A Decision Tree Based Classification Model to Automate Trip Purpose Derivation.
Terry Griffin, Yan Qun Huang · Computer Applications in Industry and Engineering · 2005
Departments of Transportation (DoT) at the state and federal level are consistently try to log travel behavior of individuals to help with urban planning. This is done by placing a Global Positioning System (GPS) in a users’ vehicle, and augmenting the electronic data with a travel diary to classify the purpose of each trip. Travel diary’s are typically kept to log the time, length, and purpose of all trips throughout an individuals day. Using just electronic data will not allow for the trip purpose to be derived, however. Unfortunately an inaccurate or incomplete travel diary threatens the accuracy of the results and ultimately any study using the data. This study presents a classification model to determine trip purpose from passively collected GPS data and eliminates the need for a travel diary altogether. This solution is useful not just because of the improvement in gaining accurate data, but it allows more data to be collected and classified by eliminating the need for active user participation in the data collection process.