Linear Support Vector

Linear Support Vector Machines (SVMs) are a powerful class of algorithms used for linear classification tasks, aiming to find the optimal hyperplane that maximizes the margin between different data classes. Current research focuses on improving efficiency through techniques like SDP decomposition for feature selection and adapting SVMs for imbalanced datasets using cost-sensitive approaches. Applications span diverse fields, including natural language processing (e.g., endangered language text classification), medical diagnosis (e.g., esophageal function analysis), and cyberbullying detection, demonstrating their broad utility and ongoing refinement. Furthermore, research emphasizes enhancing interpretability through methods like logic-based explanations and counterfactual analysis to increase trust and understanding of SVM model predictions.