AI-Based Mode of Transportation and Destination Classification and Prediction in Origin-Destination Surveys
Bibliographic record
Abstract
Travel patterns and mode choice depend on individual socio-economic attributes that need better understanding. As a result, deciding which features to investigate is a challenge in data analysis. \nThis study investigates people's activities and trips to explore the correlation between individual and household socio-economic attributes, neighbourhood socioeconomic level and land use, and the choice of mode of transportation to access destinations in the city of Montreal. The study found that the land-use characteristics of Montreal and the shapes of its residents' travel patterns impact the design and implementation of public transportation projects throughout the census agglomeration of Montreal. These transportation infrastructure influences people's commuting behaviour patterns. How to predict these patterns using historical data and existing master plans is a major goal of this work. Machine learning and deep learning algorithms were used to predict trip destination and mode of transportation. \nNumerous factors influence a person's travel pattern, including their age, residence location, and purpose of the trip. The most critical attributes were detected based on feature extraction methods and correlations between features were analyzed using a correlation heat map. This allowed to determine the most significant features to predict the trip's destination and mode of transportation. \nThree most recent versions (2008-2013-2018) of the Montreal Origin-Destination (OD) data were used. Furthermore, a comparison between the accuracy of several well-known algorithms, such as decision trees, random forests, SVMs, and feedforward neural networks, was conducted. Comparing different results yielded from different algorithms shows that neural networks outperform all the other algorithms in terms of accuracy in predicting both modes of transportation and destination (78 percent in mode choice and 68.7 percent in destination). Therefore, it was used to predict the future trip pattern of the year 2023. \nMoreover, this study proposes a Bayesian network to forecast the entire trip patterns for Montreal in 2023. This network is used to create a scaled-down version of OD2023. For this purpose, both OD and census data were used for the past 15 years. Different characteristics of trip patterns in each year were plotted. The Bayesian network captured and modelled how the trips changed over time. \nThis study provides a baseline for developing an application to extract critical statistical information about trip patterns on a neighbourhood scale in Montreal. Finally, the foundations for an application to extract critical statistical information about various trip patterns in various Montreal neighbourhoods were created. \nThis section combined various datasets from different years, including Census, land \nuse, and OD survey data. This application displays the extracted data in various plots and tables. \n \nThis research is meant to serve as a summary of previous studies as well as a reference for future research.
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 | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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".