Forecasting daily travel mode choice of kuantan travellers by means of machine learning models
In transportation studies, forecasting users’ mode choice in daily commute is crucial in order to manage traffic problems due to high number of private vehicles on the road. Conventional statistical techniques have been widely used in order to study users’ mode choice; however, the choice of the mos...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
Springer Science and Business Media Deutschland GmbH
2022
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/39753/1/Forecasting%20Daily%20Travel%20Mode%20Choice%20of%20Kuantan%20Travellers.pdf http://umpir.ump.edu.my/id/eprint/39753/2/Forecasting%20daily%20travel%20mode%20choice%20of%20kuantan%20travellers%20by%20means%20of%20machine%20learning%20models_ABS.pdf http://umpir.ump.edu.my/id/eprint/39753/ https://doi.org/10.1007/978-981-33-4597-3_89 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In transportation studies, forecasting users’ mode choice in daily commute is crucial in order to manage traffic problems due to high number of private vehicles on the road. Conventional statistical techniques have been widely used in order to study users’ mode choice; however, the choice of the most appropriate forecasting method still remains a significant concern. In this paper, we investigate the application of a number of machine learning models, namely Random Forest (RF), Tree, Naïve Bayes (NB), Logistic Regression (LR), k-Nearest Neighbour (k-NN), Support Vector Machine (SVM), as well as Artificial Neural Networks (ANN) in predicting the daily travel mode choice in Kuantan. The data was collected from a survey of Revealed/Stated Preferences (RPSP) Survey among Kuantan travellers in which eight features were taken into consideration in the present study. The classifiers were trained on the collected dataset by using five-folds cross-validation method to predict the daily mode choice. It was shown from this preliminary study that the RF, as well as ANN classifiers, could provide satisfactory classification accuracies to up to 70% in comparison to the other models evaluated. Therefore, it could be concluded that the evaluated features are rather important in deciding the travel model choice of Kuantan travellers. |
|---|