RECORD
A data-centric weak supervised learning for highway traffic incident detection
Machine Learning · Submitted 17 Dec 2021
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Overall, we show that our proposed weak supervised learning workflow achieves a high incident detection rate (0.90) and low false alarm rate (0.08).
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Using the data from loop detector sensors for near-real-time detection of traffic incidents in highways is crucial to averting major traffic congestion. While recent supervised machine learning methods offer solutions to incident detection by leveraging human-labeled incident data, the false alarm rate is often too high to be used in practice. Specifically, the inconsistency in the human labeling of the incidents significantly affects the performance of supervised learning models. To that end, we focus on a data-centric approach to improve the accuracy and reduce the false alarm rate of traffic incident detection on highways. We develop a weak supervised learning workflow to generate high-quality training labels for the incident data without the ground truth labels, and we use those generated labels in the supervised learning setup for final detection. This approach comprises three stages. First, we introduce a data preprocessing and curation pipeline that processes traffic sensor data to generate high-quality training data through leveraging labeling functions, which can be domain knowledge-related or simple heuristic rules. Second, we evaluate the training data generated by weak supervision using three supervised learning models -- random forest, k-nearest neighbors, and a support vector machine ensemble -- and long short-term memory classifiers. The results show that the accuracy of all of the models improves significantly after using the training data generated by weak supervision. Third, we develop an online real-time incident detection approach that leverages the model ensemble and the uncertainty quantification while detecting incidents. Overall, we show that our proposed weak supervised learning workflow achieves a high incident detection rate (0.90) and low false alarm rate (0.08).
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Using the data from loop detector sensors for near-real-time detection of traffic incidents in highways is crucial to averting major traffic congestion. | background 1.00 | 0.06 | 0.02 | asserted 1.00 |
| s1 | While recent supervised machine learning methods offer solutions to incident detection by leveraging human-labeled incident data, the false alarm rate is often too high to be used in practice. | background 0.66 | 0.06 | 0.02 | asserted 0.91 |
| s2 | Specifically, the inconsistency in the human labeling of the incidents significantly affects the performance of supervised learning models. | background 0.87 | 0.11 | 0.02 | asserted 0.99 |
| s3 | To that end, we focus on a data-centric approach to improve the accuracy and reduce the false alarm rate of traffic incident detection on highways. | contribution 0.90 | 0.34 | 0.05 | asserted 0.79 |
| s4 | We develop a weak supervised learning workflow to generate high-quality training labels for the incident data without the ground truth labels, and we use those generated labels in the supervised learning setup for final detection. | method 0.57 | 0.96 | 0.08 | asserted 1.00 |
| s5 | This approach comprises three stages. | method 0.96 | 0.49 | 0.01 | asserted 0.90 |
| s6 | First, we introduce a data preprocessing and curation pipeline that processes traffic sensor data to generate high-quality training data through leveraging labeling functions, which can be domain knowledge-related or simple heuristic rules. | method 0.99 | 0.93 | 0.02 | asserted 1.00 |
| s7 | Second, we evaluate the training data generated by weak supervision using three supervised learning models -- random forest, k-nearest neighbors, and a support vector machine ensemble -- and long short-term memory classifiers. | method 0.99 | 0.95 | 0.03 | asserted 0.84 |
| s8 | The results show that the accuracy of all of the models improves significantly after using the training data generated by weak supervision. | result 1.00 | 0.95 | 0.16 | asserted 1.00 |
| s9 | Third, we develop an online real-time incident detection approach that leverages the model ensemble and the uncertainty quantification while detecting incidents. | method 0.96 | 0.95 | 0.04 | asserted 0.99 |
| s10 | Overall, we show that our proposed weak supervised learning workflow achieves a high incident detection rate (0.90) and low false alarm rate (0.08). | result 1.00 | 0.97 | 0.90 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v2
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Using the data from loop detector sensors for near-real-time detection of traffic incidents in highways is crucial to averting major traffic congestion. While recent supervised machine learning methods offer solutions to incident detection by leveraging human-labeled incident data, the false alarm rate is often too high to be used in practice. Specifically, the inconsistency in the human labeling of the incidents significantly affects the performance of supervised learning models. To that end, we focus on a data-centric approach to improve the accuracy and reduce the false alarm rate of traffic incident detection on highways. We develop a weak supervised learning workflow to generate high-quality training labels for the incident data without the ground truth labels, and we use those generated labels in the supervised learning setup for final detection. This approach comprises three stages. First, we introduce a data preprocessing and curation pipeline that processes traffic sensor data to generate high-quality training data through leveraging labeling functions, which can be domain knowledge-related or simple heuristic rules. Second, we evaluate the training data generated by weak supervision using three supervised learning models -- random forest, k-nearest neighbors, and a support vector machine ensemble -- and long short-term memory classifiers. The results show that the accuracy of all of the models improves significantly after using the training data generated by weak supervision. Third, we develop an online real-time incident detection approach that leverages the model ensemble and the uncertainty quantification while detecting incidents. Overall, we show that our proposed weak supervised learning workflow achieves a high incident detection rate (0.90) and low false alarm rate (0.08).
Human review
How the literature uses it
1 arXiv paper cite this in 1 sentence
Papers citing it
Cites it as background 1 paper
-
For example, one can use traditional neural networks, such as fully connected neural networks or recurrent neural networks (RNN), for time-dependent prediction tasks [ref] .
DeepGraphONet: A Deep Graph Operator Network to Learn and Zero-shot Transfer the Dynamic Response of Networked Systems · Introduction · 0.84
What it cites
Uses the method or tool of 1 paper
-
Therefore, we used an alternative approach called deep ensemble [ref] to estimate the predictive uncertainty of our incident detection model.
Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles · Uncertainty quantification using deep ensemble · 0.98
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.