Diabetes Disease Prediction Using Various Metaheuristic Optimization Algorithms

Big data is increasingly being applied in a wide variety of fields, including industry, finance, and medicine, because of its ability to handle the hurdles of processing enormous amounts of data. The Map Reduce model is one of the data mining strategies that are frequently and efficiently used to classify huge data. In this research, a strategy for big data categorization is created using PSO, GA, ACO, FireFly, Whale, Cuckoo, and MayFly as well as a compared result of these 7 optimization classifications with 8 base classifications (RF, LR, Extra Tree, KNN, LinearSVC, GaussianNB, SVC, DT) in diabetes disease prediction. The major purpose of this work is to discover the best outcomes for diabetes illness prediction in terms of accuracy and computational time. In this project, we used seven optimization algorithms PSO, GA, ACO, FireFly, Whale, Cuckoo, and MayFly for the diabetes prediction model. We have also included the performance evaluation of the models with different evaluation metrics like precision, recall, and F1-score. We obtained an overall accuracy of 94% in the MayFly algorithm and 95.05% accuracy using the Whale Algorithm.