Granular Ball K-Class Twin Support Vector Classifier

This paper introduces the Granular Ball K-Class Twin Support Vector Classifier (GB-TWKSVC), a novel multi-class classification framework that combines Twin Support Vector Machines (TWSVM) with granular ball computing. The proposed method addresses key challenges in multi-class classification by utilizing granular ball representation for improved noise robustness and TWSVM’s non-parallel hyperplane architecture solves two smaller quadratic programming problems, enhancing efficiency. Our approach introduces a novel formulation that effectively handles multi-class scenarios, advancing traditional binary classification methods. Experimental evaluation on nine UCI benchmark datasets demonstrates that GB-TWKSVC significantly outperforms state-of-the-art classifiers in both accuracy and computational performance, achieving up to 5% higher accuracy and 50% faster computation than Twin-KSVC and 1-versus-rest TSVM. Notably, it attains 99.34% accuracy on Iris and 91.04% on Ecoli, surpassing competing methods. The method’s effectiveness is validated through comprehensive statistical tests and complexity analysis, establishing a mathematically sound framework. The results highlight GB-TWKSVC’s potential in pattern recognition, fault diagnosis and large-scale data analytics utilizing its ability to capture fine-grained features in high-dimensional data making it a valuable advancement in classification algorithms.