Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data

Abstract

Global Positioning System (GPS) technologies have been increasingly considered as an alternative to traditional travel survey methods to collect activity-travel data. Algorithms applied to extract activity-travel patterns vary from informal ad-hoc decision rules to advanced machine learning methods and have different accuracy. This paper systematically compares the relative performance of different algorithms for the detection of transportation modes and activity episodes. In particular, naive Bayesian, Bayesian network, logistic regression, multilayer perceptron, support vector machine, decision table, and C4.5 algorithms are selected and compared for the same data according to their overall error rates and hit ratios. Results show that the Bayesian network has a better performance than the other algorithms in terms of the percentage correctly identified instances and Kappa values for both the training data and test data, in the sense that the Bayesian network is relatively efficient and generalizable in the context of GPS data imputation.

Travel mode detection based on GPS track data and Bayesian networks

“…For the current study, each continuous variable was divided into three equal parts. In addition, based on the sample partition adopted by Feng and Timmermans (2014), three quarters of the sample is taken as the training set, while the remaining quarter is regarded as the test set. Consequently, the sample size of the training set and test set is 3514 and 1171, respectively.…”

The impact of Planned Special Events (PSEs) on urban traffic congestion

“…His proposal offers a panorama of system from the client's perspective and other of the server and the system is based on processes in real time and how they could be optimized algorithms and energy consumption in the smartphones. Also, Feng and Timmermans [66] presents a comparison about algorithms used to extract acceleration patterns that identify modes of transport over segments with length and time limits. His study shows a taxonomy of algorithms based on Bayesian networks, regressions linear, vector machine and decision tables.…”

Who you are is how you travel: A framework for transportation mode detection using individual and environmental characteristics

“…As a classier, it has been found to have a reduced accuracy in the context of transportation mode inference compared to other classifiers when the feature space is limited to quantities such as speed, acceleration and heading (Stenneth et al 2011;Reddy et al 2010). On the other hand, in the case the feature space is broadened with variables such as distance to metro or bus lines, Naive Bayes has been found to perform better than any discriminant classier (Feng & Timmermans 2016).…”