Comprehensive evaluation of multiple machine learning classifiers for predicting freeway incident duration

摘要

This study compares the accuracy and complexity of eleven machine learning classifiers for the problem of incident durationprediction. The proposed framework integrates feature selection and modeling techniques to evaluate the effect of multipleinfluencing factors and choose the best model for predicting incident durations. Models were developed and tested using anincident dataset collected from the Houston TranStar incidents archive, including more than 110,000 records. Features wereselected based on integrating information gain, correlation-based, and relief-based evaluators’ results. The developed andfine-tuned classifiers were compared in terms of multiple accuracy measures (precision, recall, F-1 score, and AUC) andcomplexity measures (memory storage, training time, and testing times). Overall, results showed that among the developedmodels, the support vector machines (SVM), K-Nearest Neighborhoods, and Gaussian processes classification outperformedother classifiers with a prediction accuracy of 97%. The Decision Tree classifier recorded the lowest performance with aprediction accuracy of 82%. Considering a trade-off between the model’s accuracy and complexity, the classifier with higheraccuracy associated with low training time complexity was the K-Nearest Neighborhoods achieving an accuracy of 97%,0.024 s of training time, 0.042 s of testing time, and a memory storage of 0.04 megabytes. Nevertheless, the SVM achievedthe same accuracy of 97% yet consumed much lower memory storage of 0.004 megabytes and a testing time of 0.01 s.Although the K-NN recorded the lowest training time, the SVM can be considered the best model for the ID-predictionclassification problem.