Brief Study of Classification Algorithms in Machine Learning

Abstract

Support Vector Machine (SVM) is a powerful supervised learning algorithm widely used for classification and regression tasks. It is particularly effective in high-dimensional spaces and in cases where the number of dimensions exceeds the number of samples. SVM works by finding an optimal hyperplane that maximizes the margin between different classes, ensuring better generalization to unseen data. This thesis provides a brief study of SVM, covering its fundamental principles, mathematical formulation, and different kernel functions used to handle non-linearly separable data. The study explores the advantages of SVM, such as its robustness against overfitting and its effectiveness in handling small datasets. Additionally, the limitations of SVM, including computational complexity and sensitivity to parameter selection, are discussed. Experimental analysis is conducted using real-world datasets to evaluate the performance of SVM compared to other classification techniques. The results demonstrate the impact of different kernel functions and hyperparameter tuning on classification accuracy. The findings highlight the significance of SVM in machine learning applications, making it a valuable tool in various domains, including image recognition, bioinformatics, and finance. This research aims to provide a clear understanding of SVM and its practical applications, serving as a foundation for further studies and improvements in machine learning algorithms.