Comparative Analysis of Polynomial and Radial Basis Function Models in Linear Basis Function (LBF) Models Project (Tool: Python)
- 10.2023
- Course: Data Science in Business
- Conducted rigorous hyperparameter tuning for polynomial and radial basis function models, exploring a comprehensive grid of potential values (polynomial degree p ∈ {1, 2, ..
- 10} and spread s ∈ {0.1, 0.2, ..
- 1.0}), reflecting a methodical approach to model optimization
- Employed validation set MSE for hyperparameter optimization, followed by a comparative analysis of model performances using test set MSE, indicating a robust approach to model evaluation
- Effectively utilized data visualization to present model predictions and performance comparisons, providing intuitive insights into the adaptability and efficacy of polynomial and radial basis function models in handling diverse datasets
- Revealed crucial findings on the suitability and performance differences between polynomial and radial basis functions, emphasizing the importance of informed model selection and hyperparameter optimization in enhancing prediction accuracy and efficiency
- Demonstrated in-depth analysis of four datasets encompassing 20,000 observations to unravel complex data patterns, showcasing strong analytical skills
