An Unsupervised Feature Learning Approach to Improve Automatic Incident Detection
This work addresses the need for better feature representation in incident detection for transportation systems, offering an incremental improvement by applying unsupervised learning to reduce reliance on labeled data.
The paper tackled the problem of improving automatic incident detection in transportation systems by using an unsupervised feature learning algorithm to generate higher-level features, resulting in significant improvements in detection rate, false alarm rate, and mean time to detect across three representative cases.
Sophisticated automatic incident detection (AID) technology plays a key role in contemporary transportation systems. Though many papers were devoted to study incident classification algorithms, few study investigated how to enhance feature representation of incidents to improve AID performance. In this paper, we propose to use an unsupervised feature learning algorithm to generate higher level features to represent incidents. We used real incident data in the experiments and found that effective feature mapping function can be learnt from the data crosses the test sites. With the enhanced features, detection rate (DR), false alarm rate (FAR) and mean time to detect (MTTD) are significantly improved in all of the three representative cases. This approach also provides an alternative way to reduce the amount of labeled data, which is expensive to obtain, required in training better incident classifiers since the feature learning is unsupervised.