Comparison of Classification Methods for Golf Putting Performance Analysis

This paper presents a comparative case study on the classification accuracy between 5 methods for golf putting performance analysis. In a previous work, a digital camera was used to capture 30 trials of 6 expert golf players. The detection of the horizontal position of the golf club was performed using a computer vision technique followed by the estimation algorithm Darwinian Particle Swarm Optimization (DPSO) in order to obtain a kinematical model of each trial. In this paper, the estimated parameters of the models are used as sample and training data of five classification algorithms: 1) Linear Discriminant Analysis (LDA); 2) Quadratic Discriminant Analysis (QDA); 3) Naïve Bayes with Normal (Gaussian) distribution (NV); 4) Naïve Bayes with Kernel Smoothing Density Estimate (NVK) and 5) Least Squares Support Vector Machines with Radial Basis Function Kernel (LS-SVM). The 5 classification methods are then compared through the analysis of the confusion matrix and the area under the Receiver Operating Characteristic curve (AUC).