Prediction of Individual Social-Demographic Role Based on Travel Behavior Variability Using Long-Term GPS Data

Abstract

With the development of and advances in smartphones and global positioning system (GPS) devices, travelers' long-term travel behaviors are not impossible to obtain. This study investigates the pattern of individual travel behavior and its correlation with social-demographic features. For different social-demographic groups (e.g., full-time employees and students), the individual travel behavior may have specific temporal-spatial-mobile constraints. The study first extracts the home-based tours, including Home-toHome and Home-to-Non-Home, from long-term raw GPS data. The travel behavior pattern is then delineated by home-based tour features, such as departure time, destination location entropy, travel time, and driving time ratio. The travel behavior variability describes the variances of travelers' activity behavior features for an extended period. After that, the variability pattern of an individual's travel behavior is used for estimating the individual's social-demographic information, such as social-demographic role, by a supervised learning approach, support vector machine. In this study, a long-term (18-month) recorded GPS data set from Puget Sound Regional Council is used. The experiment's result is very promising. The sensitivity analysis shows that as the number of tours thresholds increases, the variability of most travel behavior features converges, while the prediction performance may not change for the fixed test data.

A Deep Learning Approach to Infer Employment Status of Passengers by Using Smart Card Data

“…After extracting temporal profiles of individuals, the employment status can be inferred using classifiers. A wide range of traditional classifiers has been applied for performing the classification task, including support vector machine, decision tree, and Naïve Bayes [8], [37]. Such classifiers usually need hand-crafted features as input for training.…”

“…To determine the best hyperparameters, the basic way is to change one of the parameters while the other parameters remain unchanged, which is called a grid search. For each channel of the TMC-CNN, the number of a combination of the convolutional and average pooling layers was chosen from 1 to 4, and each layer's number of filters was selected from [4], [8], [16], [32]. In addition, to the best of our knowledge, there is no rule of thumb to choose the size of the filters.…”

“…It is useful for transport planning, urban design, and personalized service [7], [8]. Among various socio-demographic characteristics, the ability to characterize the employment status of passengers based on SCD has numerous applications in modern PT systems and other areas.…”

“…RELATED WORK Employment status is a key demographic characteristic for travel behavior studies. Several works have attempted to infer the demographics from human mobility by using GPS trajectories [8], [16] or check-ins [17]. Both data sources can reveal the frequently visited locations, which appear a high degree of temporal and spatial regularity with significant probability [18].…”

“…Both data sources can reveal the frequently visited locations, which appear a high degree of temporal and spatial regularity with significant probability [18]. For example, Zhu et al [8] used an 18-month GPS dataset of 275 households to predict individuals' socialdemographic information, including age group, gender, and employment status. Home-based tours were first detected from GPS trajectories and features and variability were then extracted, which were fed into the Support Vector Machine (SVM) and logistic regression models.…”