Automatic incident detection algorithm based on under-sampling for imbalanced traffic data

The rapid increase in the economy and ownership of car has resulted in the increased growth of demand of transportation system. As a result, traffic operation conditions deteriorate and traffic incidents occur in an increasing frequency. Unexpected traffic incidents, especially secondary traffic incidents usually result in undesirable traffic congestion and higher probability of severe accidents. Automatic Incident Detection (AID), as an important component of Nation’s Intelligent Transportation Systems (ITS) (JEONG, Y, et al., 2011), has great significance on reducing the delays and congestion.