Published Paper Details:

Abstract-- According to the World Health Organization (WHO), there are 420 million people affected with diabetics worldwide, and this has caused an increase in the mortality rate to nearly one million per year. A serious situation has resulted from this exceptional rise in cases and fatalities, since the data statistics show a considerable rise in diabetic cases amid youngsters. For accurate diabetes categorization and prediction, this study makes use of a new framework called Binary Chaotic Optimized Fused Learning (BCoFL). In this case, feature selection and data dimensionality reduction are accomplished using the Binary Chaotic Hunger Game Search (BCHGS) optimization technique. For system validation and evaluation, the diabetes dataset from PIMA Indian patients was employed. Data cleaning and normalization processes are first used to preprocess the dataset. Additionally, the disease is predicted from the supplied data effectively and with little overfitting using the Fused Learning Classification Algorithm (FLCA). The performance and prediction results of the suggested CoFL technique are validated and compared using the accuracy, precision, prediction rate, and other parameters.