Supervised Land Use Inference from Mobility Patterns
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
This paper addresses the relationship between land use and mobility patterns. Since each particular zone directly feeds the global mobility once acting as origin of trips and others as destination, both roles are simultaneously used for predicting land uses. Specifically this investigation uses mobility data derived from mobile phones, a technology that emerges as a useful, quick data source on people’s daily mobility, collected during two weeks over the urban area of Malaga (Spain). This allows exploring the relevance of integrating weekday-weekend trip information to better determine the category of land use. First, this work classifies patterns on trips originated and terminated in each zone into groups by means of a clustering approach. Based on identifiable relationships between activity and times when travel peaks appear, a preliminary categorization of uses is provided. Then, both grouping results are used as input variables in a K -nearest neighbors (KNN) classification model to determine the exact land use. The KNN method assumes that the category of an object must be similar to the category of the closest neighbors. After training the models, the findings reveal that this approach provides a precise land use categorization, yielding the best accuracy results for the major categories of land uses in the studied area. Moreover, as a result, the weekend data certainly contributes to finding more precise land uses as those obtained by just weekday data. In particular, the percentage of correctly predicted categories using both weekday and weekend is around 80%, while just weekday data reach 67%. The comparison with actual land uses also demonstrates that this approach is able to provide useful information, identifying zones with a specific clear dominant use (residential, industrial, and commercial), as well as multiactivity zones (mixed). This fact is especially useful in the context of urban environments where multiple activities coexist.
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.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 it