Support Vector Regression

Support Vector Regression (SVR) is a machine learning technique used for regression tasks, aiming to find the optimal hyperplane that best fits a dataset while minimizing prediction error. Current research emphasizes enhancing SVR's accuracy and robustness through hybrid models combining SVR with meta-heuristic optimization algorithms (like Particle Swarm Optimization and Grey Wolf Optimization) for hyperparameter tuning, and exploring novel loss functions to improve resilience to outliers and noise. These advancements are driving applications across diverse fields, including air quality forecasting, 3D point cloud quality assessment, and various engineering and scientific prediction problems, demonstrating SVR's versatility and practical impact.

Papers