Disaggregate travel demand analysis using big data sources: unsupervised learning methods for data-driven trip purpose estimation
Bibliographic record
Abstract
Data is a paramount factor in the success of transport modeling. Smartphones can be employed to retrieve full individual trajectories, either locally through apps using the devices’ integrated GPS sensors, or through the mobile network operator (MNO), by tracing the mobile antennas to which devices connect over time. Several studies have demonstrated of the utility of this data to infer users’ door-to-door trips, and then to build door-to-door origin-destination matrices, which are a key feedstock for transport modeling and planning. Some MNOs already provide such services commercially, yielding notable time and cost savings with respect to matrices estimated through traditional surveys. However, these are often highly aggregated and lack supplementary relevant trip information, such as mode and purpose, and if any, these are commonly obtained by means of human-driven heuristic considerations and fixed rules. This study aims at exploring the suitability of machine learning techniques for data-driven mobility demand estimation and analysis. It identifies associated opportunities and challenges through a pilot experiment focused on trip purpose estimation via diverse clustering techniques. Despite the experiment's limitations due to a small sample size and altered mobility patterns resulting from the COVID-19 pandemic, clustering algorithms (both distance- and density-based) successfully yield meaningful outcomes. The results include the identification of travel purposes, such as trips to home with or without overnight stays, trips to occasional destinations, commutes to work, trips to holiday stays, and more. These preliminary yet promising findings suggest that machine learning holds significant potential in mobility analysis, and it could feasibly be employed to estimate big-data-driven demand matrices, offering a higher degree of disaggregation and consequently enhancing the quality of transport modeling practices.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.005 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".