Classification of User Preference for Self-Driving Mode and Behaviors of Autonomous Vehicle
Bibliographic record
Abstract
In the realm of semi-autonomous vehicles (semi-AVs), diverse facets of driver preferences in human-driver interaction have been explored. This study delves into the pivotal aspect of whether drivers favor a self-driving mode within semi-AVs. To discern such preferences, we conducted a user preference survey and leveraged the collected data to construct machine learning (ML) models capable of classifying these preferences effectively. Focusing on the prediction of preferred self-driving actions, we discerned user preferences at four granularity levels. At the lowest level, we ascertained whether users leaned towards self-driving mode at each traffic situation. For the highest granularity, we identified five distinct action types, each comprising two-staged actions (‘Act-Inform,’ ‘Inform-Act,’ ‘Act-No Inform,’ ‘Inform-Consent,’ ‘Alert-Handover’). Our online survey involved 85 participants from each age group (23-44 and 60+), who responded to a background questionnaire and situation-based inquiries for eighteen selected traffic situations. These situations were recorded in two regions of Ontario, Canada (Toronto and Waterloo), representing different population sizes and traffic conditions. Responses from the online survey were processed into features for ML models of user preference prediction. ML model optimization was achieved through Bayesian hyperparameter optimization and the Boruta SHAP (SHapley Additive exPlanations) feature selection algorithm. Boruta SHAP evaluated features and highlighted important features in each ML model of the two age group models (23-44 and 60+). Our findings underscore the feasibility of developing predictive models for driver preferences in self-driving behaviors, with average accuracies exceeding 85% and 72% at the lowest and highest granularity levels, respectively.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), 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".