Adaptive Cluster-Based Synthetic Minority Oversampling Technique for Traffic Mode Choice Prediction with Imbalanced Dataset

TL;DR

The paper tackles imbalanced transportation mode data by introducing AC-SMOTE, an adaptive cluster-based oversampling technique that uses DBSCAN to identify homogeneous subgroups within minority classes and oversamples within each cluster. When paired with random forest or XGBoost, AC-SMOTE improves minority class F1 scores and maintains performance on majority classes, achieving notable ROC-AUC values (e.g., 0.94 for Walking and 0.84 for Cycling) while highlighting ongoing challenges for highly underrepresented classes like Cycling. The methodology is validated on the LPMC dataset and shows potential generalizability to other urban contexts, contributing to more accurate and equitable traffic mode predictions for planning and resource allocation. The work suggests directions for further enhancement through advanced clustering, feature engineering, and testing on newer mobility modalities.

Abstract

Urban datasets such as citizen transportation modes often contain disproportionately distributed classes, posing significant challenges to the classification of under-represented samples using data-driven models. In the literature, various resampling methods have been developed to create synthetic data for minority classes (oversampling) or remove samples from majority classes (undersampling) to alleviate class imbalance. However, oversampling approaches tend to overgeneralize minor classes that are closely clustered and neglect sparse regions which may contain crucial information. Conversely, undersampling methods potentially remove useful information on certain subgroups. Hence, a resampling approach that takes the inherent distribution of data into consideration is required to ensure appropriate synthetic data creation. This study proposes an adaptive cluster-based synthetic minority oversampling technique. Density-based spatial clustering is applied on minority classes to identify subgroups based on their input features. The classes in each of these subgroups are then oversampled according to the ratio of data points of their local cluster to the largest majority class. When used in conjunction with machine learning models such as random forest and extreme gradient boosting, this oversampling method results in significantly higher F1 scores for the minority classes compared to other resampling techniques. These improved models provide accurate classification of transportation modes.

Figures (3)

• Figure 1: The distribution of the travel modes in the dataset.
• Figure 2: DBScan clustering of "Public Transport," "Walking," and "Cycling" classes.
• Figure 3: Synthetic data creation for the minority classes.