CP3

Discovering Traveling Companions using Autoencoders (2020): Proposes an autoencoder-based representation learner and then clusters embeddings (example; DBSCAN) to find companion groups. It provides strong baseline for “unsupervised companion discovery.”
Graph-based representation for identifying individual travel activities with spatiotemporal trajectories and POI data (2022): It constructs a weighted directed mobility graph over activity zones and learns embeddings (Gstp2Vec) for downstream identification tasks. This is the main reference for how to build a graph representation from trajectories/POIs.

Offline map matching using time-expanded graph for low-frequency data (2021): It explicitly constructs a time-expanded graph where a source to sink path is a candidate route. It motivates time-expanded representations as robust for low-frequency/noisy trajectories.

Towards geographically robust statistically significant regional colocation pattern detection (2022): This paper highlights that regional co-location mining without significance testing yields spurious patterns. It also introduces a significance-based mining approach using geographically defined regions.
Discovering Super-Colocation Patterns: A Summary of Results (2025): It extends co-location discovery toward “super-colocation” (higher-order) patterns. which is a good motivation for moving from pairs to groups.
Reducing False Discoveries in Statistically-Significant Regional-Colocation Mining: A Summary of Results (2024): This paper is important for “false discovery control” as a core issue when defining co-occurrence patterns

Exploring human mobility: a time-informed approach to pattern mining and sequence similarity (2025): This paper introduces time-informed pattern mining and time-aware sequence similarity.

CoBAD (2025): It formulates collective mobility as collective event sequences with a co-occurrence event graph, then uses masked event/link reconstruction to learn interactions. It should be helpful as a “modern graph pretraining & link reconstruction” reference.

Travel Trajectory Frequent Pattern Mining Based on Differential Privacy Protection (2021): This paper shows frequent pattern mining under differential privacy constraints. It can be cited as adjacent work if we mention privacy or data sharing constraints.

Discovery of Travelling Companions from trajectories with different sampling rates (2020): This paper directly supports our problem statement that co-travel definitions must handle sampling mismatch it is also useful for robustness evaluation framing.

2. Decision Table:

Table

Page updated

Report abuse