GA-KELM: Genetic-Algorithm-Improved Kernel Extreme Learning Machine for Traffic Flow Forecasting

Abstract

A prompt and precise estimation of traffic conditions on the scale of a few minutes by analyzing past data is crucial for establishing an effective intelligent traffic management system. Nevertheless, because of the irregularity and nonlinear features of traffic flow data, developing a prediction model with excellent robustness poses a significant obstacle. Therefore, we propose genetic-search-algorithm-improved kernel extreme learning machine, termed GA-KELM, to unleash the potential of improved prediction accuracy and generalization performance. By substituting the inner product with a kernel function, the accuracy of short-term traffic flow forecasting using extreme learning machines is enhanced. The genetic algorithm evades manual traversal of all possible parameters in searching for the optimal solution. The prediction performance of GA-KELM is evaluated on eleven benchmark datasets and compared with several state-of-the-art models. There are four benchmark datasets from the A1, A2, A4, and A8 highways near the ring road of Amsterdam, and the others are D1, D2, D3, D4, D5, D6, and P, close to Heathrow airport on the M25 expressway. On A1, A2, A4, and A8, the RMSEs of the GA-KELM model are 284.67 vehs/h, 193.83 vehs/h, 220.89 vehs/h, and 163.02 vehs/h, respectively, while the MAPEs of the GA-KELM model are 11.67%, 9.83%, 11.31%, and 12.59%, respectively. The results illustrate that the GA-KELM model is obviously superior to state-of-the-art models.

Lag‐related noise shrinkage stacked LSTM network for short‐term traffic flow forecasting

“…These methods include the exponential smoothing model (ES), grey model (GM), least-squares boosting (LSBOOST), support vector regression method (SVR), stacked autoencoder model (SAE), Kalman filtering model (KF), and LSTM model. Besides, six latest stateof-the-art models are included, SVRGSAS [19], SrOrkNNr [46], GA-KELM [21], PSOGSA-ELM [47], ABC-ELM [48],…”

“…The emerging machine learning methods have been employed for short-term traffic flow forecasting, such as deep belief network [14], fuzzy logic [15], Kalman filter [16,17], ensemble learning [18], support vector regression [19], k-nearest neighbor [20], and extreme learning machines [21]. Machine learning methods require a large amount of sample data and sufficient training effort to establish the mapping function [22,23].…”

Network-Level Hierarchical Bottleneck Congestion Control Method for a Mixed Traffic Network

“…It is well-suited for implementation in distributed, cloud, or multi-core computing environments, thus expediting the whole solution process. Furthermore, it has been observed that GA can work effectively in complicated contexts and is robust to the presence of noise, uncertainty, and non-convex functions in the problem [47][48][49].…”

AST3DRNet: Attention-Based Spatio-Temporal 3D Residual Neural Networks for Traffic Congestion Prediction

“…In addition, the extreme learning machine (ELM) [ 28 , 29 , 30 ] is also a typical method that can be applied to traffic forecasting. Chai et al [ 31 ] proposed the GA-KELM to forecast traffic flow. This method reduces the overfitting problem and achieves more accurate prediction performance by optimizing the kernel limit-learning machine with a genetic algorithm.…”