Melvin Wong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Melvin Wong has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Bilal Farooq
Toronto Metropolitan University
3 shared publications - Guoyang Qin
Department of Traffic Engineering and the Key Laboratory of Road and Traffic Engineering, Ministry of Education, Tongji University, Shanghai, China
3 shared publications - Jian Sun
Tongji University
3 shared publications Chaoteng Wu
2 shared publications- Shuhan Qiu
Key Laboratory of Road and Traffic Engineering, Ministry of Education, Shanghai, China
2 shared publications - Soora Rasouli
Eindhoven University of Technology
2 shared publications
Latest Publications
- Open Access
- 31 October 2026
- Published by Springer Nature
- p. 426-434
Toward a User-Centered Privacy Framework for Transport Data Sharing: A Design Framework for Trust, Control, and Participation
Traveler data is increasingly essential for the operation of personalized mobility services, enabling integration, optimization, and tailored travel experiences. However, the governance of such data remains insufficiently operationalized, particularly in how privacy is handled across policy, corporate practice, and user perspectives. This paper draws on a review of transport policy frameworks and mobility citizen-oriented studies, combined with an assessment of transport company privacy policies against General Data Protection Regulation (GDPR) rights, to examine how mobility data sharing is currently framed and implemented. The analysis reveals a systematic misalignment across domains: (a) policy frameworks emphasize formal rights but provide limited guidance for context-specific data use, (b) companies largely comply with regulatory requirements while reducing transparency and diffusing responsibility in practice, and (c) citizen perspectives remain only partially informed, with existing research largely overlooking the role of collective benefits in shaping data-sharing decisions. By framing mobility data sharing as a governance, legal, and design challenge, this paper provides a structured understanding of current limitations in how privacy is conceptualized and operationalized in mobility systems. This foundation lays the groundwork for future empirical studies of how travelers evaluate privacy-related trade-offs.
- 31 October 2026
- Published by Springer Nature
- p. 426-434
Open AccessToward a User-Centered Privacy Framework for Transport Data Sharing: A Design Framework for Trust, Control, and Participation
Traveler data is increasingly essential for the operation of personalized mobility services, enabling integration, optimization, and tailored travel experiences. However, the governance of such data remains insufficiently operationalized, particularly in how privacy is handled across policy, corporate practice, and user perspectives. This paper draws on a review of transport policy frameworks and mobility citizen-oriented studies, combined with an assessment of transport company privacy policies against General Data Protection Regulation (GDPR) rights, to examine how mobility data sharing is currently framed and implemented. The analysis reveals a systematic misalignment across domains: (a) policy frameworks emphasize formal rights but provide limited guidance for context-specific data use, (b) companies largely comply with regulatory requirements while reducing transparency and diffusing responsibility in practice, and (c) citizen perspectives remain only partially informed, with existing research largely overlooking the role of collective benefits in shaping data-sharing decisions. By framing mobility data sharing as a governance, legal, and design challenge, this paper provides a structured understanding of current limitations in how privacy is conceptualized and operationalized in mobility systems. This foundation lays the groundwork for future empirical studies of how travelers evaluate privacy-related trade-offs.
- Open Access
- 11 February 2026
- Published by Elsevier in SSRN Electronic Journal
Lost choice sets reclaimed: A cognitively-inspired choice learning framework for refining link-based route choice modeling
Route choice modeling (RCM) aims to explain and predict how travelers select routes in transportation networks. A central challenge is defining the effective choice set -- the subset of alternatives that meaningfully influence observed decisions between an origin and a destination. In real-world networks with many structurally feasible routes, explicit path-based methods struggle to construct high-quality choice sets without relying on rigid, handcrafted rules. Link-based implicit models, such as recursive logit, avoid explicit enumeration by embedding route availability within the estimation process, offering greater flexibility. However, this implicit treatment may assign non-negligible probability to weakly relevant or behaviorally implausible routes, complicating interpretation and preference estimation. To address this trade-off, we introduce a hybrid, bi-level link-based framework that regulates the effective choice set within an implicit modeling paradigm. The proposed adaptive prism recursive logit model combines a smooth, OD-informed filtering layer that modulates route availability with a recursive logit-based evaluation layer that performs compensatory utility-based choice modeling. This separation is motivated as a modeling abstraction that enables explicit control over route screening while preserving probabilistic consistency and end-to-end estimation. Across controlled experiments, real-world networks, and ablation studies, we show that separating filtering from evaluation improves predictive accuracy, stabilizes utility estimation, and yields more compact and interpretable effective choice sets compared to standard and extended recursive logit baselines. These results demonstrate that structured regulation of implicit route sets can enhance the practical usefulness of link-based route choice models for estimation and behavioral analysis.
- 11 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessLost choice sets reclaimed: A cognitively-inspired choice learning framework for refining link-based route choice modeling
Route choice modeling (RCM) aims to explain and predict how travelers select routes in transportation networks. A central challenge is defining the effective choice set -- the subset of alternatives that meaningfully influence observed decisions between an origin and a destination. In real-world networks with many structurally feasible routes, explicit path-based methods struggle to construct high-quality choice sets without relying on rigid, handcrafted rules. Link-based implicit models, such as recursive logit, avoid explicit enumeration by embedding route availability within the estimation process, offering greater flexibility. However, this implicit treatment may assign non-negligible probability to weakly relevant or behaviorally implausible routes, complicating interpretation and preference estimation. To address this trade-off, we introduce a hybrid, bi-level link-based framework that regulates the effective choice set within an implicit modeling paradigm. The proposed adaptive prism recursive logit model combines a smooth, OD-informed filtering layer that modulates route availability with a recursive logit-based evaluation layer that performs compensatory utility-based choice modeling. This separation is motivated as a modeling abstraction that enables explicit control over route screening while preserving probabilistic consistency and end-to-end estimation. Across controlled experiments, real-world networks, and ablation studies, we show that separating filtering from evaluation improves predictive accuracy, stabilizes utility estimation, and yields more compact and interpretable effective choice sets compared to standard and extended recursive logit baselines. These results demonstrate that structured regulation of implicit route sets can enhance the practical usefulness of link-based route choice models for estimation and behavioral analysis.
- Open Access
- 20 January 2026
- Published by Elsevier in Transportation Research Part A: Policy and Practice
- Vol. 205
Understanding crowd-shipping acceptance: A hierarchical latent class personality-attitude hybrid choice model
- 20 January 2026
- Published by Elsevier in Transportation Research Part A: Policy and Practice
- Vol. 205
Open AccessUnderstanding crowd-shipping acceptance: A hierarchical latent class personality-attitude hybrid choice model
- Open Access
A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data
The emergence of data-driven demand analysis has led to the increased use of generative modelling to learn the probabilistic dependencies between random variables. Although their apparent use has mostly been limited to image recognition and classification in recent years, generative machine learning algorithms can be a powerful tool for travel behaviour research by replicating travel behaviour by the underlying properties of data structures. In this paper, we examine the use of generative machine learning approach for analyzing multiple discrete-continuous (MDC) travel behaviour data. We provide a plausible perspective of how we can exploit the use of machine learning techniques to interpret the underlying heterogeneities in the data. We show that generative models are conceptually similar to the choice selection behaviour process through information entropy and variational Bayesian inference. Without loss of generality, we consider a restricted Boltzmann machine (RBM) based algorithm with multiple discrete-continuous layers, formulated as a variational Bayesian inference optimization problem. We systematically describe the proposed machine learning algorithm and develop a process of analyzing travel behaviour data from a generative learning perspective. We show parameter stability from model analysis and simulation tests on an open dataset with multiple discrete-continuous dimensions from a data size of 293,330 observations. For interpretability, we derive the conditional probabilities, elasticities and perform statistical analysis on the latent variables. We show that our model can generate statistically similar data distributions for travel forecasting and prediction and performs better than purely discriminative methods in validation. Our results indicate that latent constructs in generative models can accurately represent the joint distribution consistently on MDC data.
Open AccessA bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data
The emergence of data-driven demand analysis has led to the increased use of generative modelling to learn the probabilistic dependencies between random variables. Although their apparent use has mostly been limited to image recognition and classification in recent years, generative machine learning algorithms can be a powerful tool for travel behaviour research by replicating travel behaviour by the underlying properties of data structures. In this paper, we examine the use of generative machine learning approach for analyzing multiple discrete-continuous (MDC) travel behaviour data. We provide a plausible perspective of how we can exploit the use of machine learning techniques to interpret the underlying heterogeneities in the data. We show that generative models are conceptually similar to the choice selection behaviour process through information entropy and variational Bayesian inference. Without loss of generality, we consider a restricted Boltzmann machine (RBM) based algorithm with multiple discrete-continuous layers, formulated as a variational Bayesian inference optimization problem. We systematically describe the proposed machine learning algorithm and develop a process of analyzing travel behaviour data from a generative learning perspective. We show parameter stability from model analysis and simulation tests on an open dataset with multiple discrete-continuous dimensions from a data size of 293,330 observations. For interpretability, we derive the conditional probabilities, elasticities and perform statistical analysis on the latent variables. We show that our model can generate statistically similar data distributions for travel forecasting and prediction and performs better than purely discriminative methods in validation. Our results indicate that latent constructs in generative models can accurately represent the joint distribution consistently on MDC data.
- Open Access
Information processing constraints in travel behaviour modelling: A generative learning approach
Travel decisions tend to exhibit sensitivity to uncertainty and information processing constraints. These behavioural conditions can be characterized by a generative learning process. We propose a data-driven generative model version of rational inattention theory to emulate these behavioural representations. We outline the methodology of the generative model and the associated learning process as well as provide an intuitive explanation of how this process captures the value of prior information in the choice utility specification. We demonstrate the effects of information heterogeneity on a travel choice, analyze the econometric interpretation, and explore the properties of our generative model. Our findings indicate a strong correlation with rational inattention behaviour theory, which suggest that individuals may ignore certain exogenous variables and rely on prior information for evaluating decisions under uncertainty. Finally, the principles demonstrated in this study can be formulated as a generalized entropy and utility based multinomial logit model.
Open AccessInformation processing constraints in travel behaviour modelling: A generative learning approach
Travel decisions tend to exhibit sensitivity to uncertainty and information processing constraints. These behavioural conditions can be characterized by a generative learning process. We propose a data-driven generative model version of rational inattention theory to emulate these behavioural representations. We outline the methodology of the generative model and the associated learning process as well as provide an intuitive explanation of how this process captures the value of prior information in the choice utility specification. We demonstrate the effects of information heterogeneity on a travel choice, analyze the econometric interpretation, and explore the properties of our generative model. Our findings indicate a strong correlation with rational inattention behaviour theory, which suggest that individuals may ignore certain exogenous variables and rely on prior information for evaluating decisions under uncertainty. Finally, the principles demonstrated in this study can be formulated as a generalized entropy and utility based multinomial logit model.
- 1 December 2018
- Published by Elsevier in Journal of Choice Modelling
- Vol. 29, 152-168
Discriminative conditional restricted Boltzmann machine for discrete choice and latent variable modelling
- 1 December 2018
- Published by Elsevier in Journal of Choice Modelling
- Vol. 29, 152-168
Discriminative conditional restricted Boltzmann machine for discrete choice and latent variable modelling