Improving SVM performance for type II diabetes prediction with an improved non-linear kernel: Insights from the PIMA dataset

•Finding limitations of traditional SVM kernels in handling complex, non-linear relationships present in medical datasets.•Developing an improved non-linear kernel for SVM in Type-II diabetes prediction.•Superior performance of the proposed method compared to traditional SVM kernels.•Describes the PIMA dataset, including its characteristics, attributes, and preprocessing steps.