Framework for Prediction of Diabetes using FireFly Swarm Intelligence Algorithm, Fuzzy C Mean and SVM Algorithm

Diabetes is a metabolic sickness that remains to be a major universal problem because it influences the health of the entire population. Over the years several researchers have attempted to build a model for predicting diabetes accurately. However, this field study remains a challenge because of the unavailability of dataset and prediction models which forces the researchers to utilize ML (Machine Learning) algorithm and Big data analytics. The primary objective of the study was to identify how ML and Big data analytics can be adopted in diabetes. The evaluation of outcomes demonstrates that the proposed ML-based technique might achieve 8.6% of accuracy. The authors performed a review of literature on the ML models and came out with the suggestion of an Automatic Intelligent method for the prediction of diabetes based on the findings. The authors explored ML approaches and proposed and developed an intelligent ML-based architecture for diabetes prediction. In this work, the authors used fuzzy mean, SVM learning models for the prediction of diabetes. A firefly algorithm optimized classifies was utilized for selecting the best features. It is proposed in the work, that an exclusive automatic intelligent optimize diabetic prediction model is built by utilizing ML approaches. The framework was created following a thorough analysis of existing prediction approaches in the literature and assessing their relevance to diabetes. The authors presented the training methods, model evaluation strategies, and the challenges allied to the prediction of diabetes also the remedies they provide by using the framework. Health professionals, students, researchers, and stakeholders working in diabetes forecast research development may benefit from the findings of the study. The proposed work achieves an accuracy of 86 percent with low error rates.