Stay Time Prediction for Individual Stay Behavior

Abstract

Stay time is important for understanding people's travel behavior and mobility motivation. In this paper, by leveraging private car trajectory data, we propose a novel systematic approach for implementing stay behavior detection and stay time prediction. Specifically, we first propose a fuzzy logicbased stay detection method for detecting stay behavior in a large-scale private car trajectory dataset. Then, we design a spatiotemporal feature extraction method called clustering and kernel (CaK) by considering the spatial similarity, temporal periodicity and spatiotemporal correlation of stay behavior data. Furthermore, we propose a stay time predictor (STP) based on gradient-boosting regression trees and a long short-term memory network that can estimate the future durations of private car users' stays in various scenarios. We perform extensive experiments based on two real-life trajectory datasets. The experimental results demonstrate that the STP achieves a predictive accuracy (specifically, the root-mean-square error) of 123.94 and R 2 of 0.893 for stay time prediction of individual stays. This study provides a new perspective for understanding people's stay behavior.INDEX TERMS Machine learning, data mining, regression analysis, prediction methods. I. INTRODUCTION A. BACKGROUND AND MOTIVATION

Analyzing the Features of Passenger Drop-Off Behavior at Airport Curbsides: A Case Study From Guangxi Province, China

“…Some researchers used statistical methods and Poisson and negative binomial regressions to analyze the impact of various factors on urban roadside parking maneuvering time [26]. Some researchers proposed the use of private car trajectory data to achieve fuzzy logic-based dwell behavior detection and dwell time inference [27]. Some proposed an airport curbside simulation model that calculates the optimal length of an airport curbside based on traffic characteristics, such as speed, dwell time, and parking demand [1].…”

An attention-based recurrent learning model for short-term travel time prediction

“…As a result, we set hidden units to 32 in our experiments. To avoid overfitting, a normalization term is added in loss computation as shown in Eq (12).…”

“…Research groups and companies analyze this data using big data and machine learning to improve people’s living standards [ 5 ]. Researchers have used data from multiple sources to improve traffic-related operations with applications in traffic congestion prediction [ 6 ], traffic flow prediction [ 7 ], traffic speed estimation [ 8 ], traffic demand prediction [ 9 ], traffic signal control [ 10 ], parking space forecasting [ 11 ], stay point detection [ 12 ], traffic accident prediction [ 13 ], accident severity analysis [ 14 ], and many others.…”