What question did this study set out to answer?

Develop a deep learning algorithm, MogLSTM-TrajGAN, to generate synthetic trajectory data that preserves location privacy while maintaining high data utility.

December 13, 2024

Deep Learning-Based Synthetic Trajectory Generation for Enhanced Privacy and Utility

View Full Paper

Ask AI

Bookmark

Authors

YZYinghui ZhangBerlin School of Economics and Law JZJuanru ZhangInner Mongolia University HLHao LiChinese Academy of Sciences

Discussion

Member takes

Implication

Simulation study reveals high-utility synthetic trajectory generation using MogLSTM-TrajGAN, indicating an effective balance between location privacy and spatio-temporal data utility.

Key Points

Develop a deep learning algorithm, MogLSTM-TrajGAN, to generate synthetic trajectory data that preserves location privacy while maintaining high data utility.
Encoded empirical trajectory data to capture underlying spatio-temporal distributions and integrated extra gating units into long short-term memory networks.
Combined the modified LSTM structure with a generative adversarial network to create synthetic trajectory sequences.
Designed a trajectory loss function to measure and constrain similarity loss between generated and empirical trajectory profiles.
Simulations showed generated trajectories retained spatial, temporal, and point-of-interest characteristics of real-world movement paths.
The algorithm achieved an effective operational balance between preserving user privacy and providing usable trajectory utility without excessive deviation.

Cite This Study

Zhang et al. (2024) studied this question.

synapsesocial.com/papers/6a15f789748d4b09ae522a30 https://doi.org/10.1109/hpcc64274.2024.00181

View Full Paper

Ask AI

Bookmark

Implication

Simulation study reveals high-utility synthetic trajectory generation using MogLSTM-TrajGAN, indicating an effective balance between location privacy and spatio-temporal data utility.

Deep Learning-Based Synthetic Trajectory Generation for Enhanced Privacy and Utility

Discussion

Member takes

Implication

Key Points

Cite This Study

Deep Learning-Based Synthetic Trajectory Generation for Enhanced Privacy and Utility

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider