Portal de Periódicos da CAPES

Framework for Diabetes Prediction Using Machine Learning Techniques Through Swarm Intelligence

2022; Springer Nature; Linguagem: Inglês

10.1007/978-981-19-0332-8_33

2524-7573

C. Kalpana, B. Booba,

Imbalanced Data Classification Techniques

Diabetes is treated as a chronic and a deadly disease across the globe. It curtails life anticipation and makes people more susceptible to cardiovascular disease. An effective diabetes prediction can help people take effective preventive measures. Medical data is complicated and unstructured, making it difficult to accurately forecast disease. However, much research has been conducted on diabetes-prediction, it remains a big challenge. The aim of the work is to state the challenge and develop a machine learning model. In this work, different attributes like BMI, Age, Blood pressure, Blood sugar, and so on has been used for diagnosing diabetes. Several machine learning techniques Support Vector Machine (SVM), Naïve Bayes, XGBoost were deployed to predict diabetes. Further, the ML algorithms were optimized by applying the Binary particle Swarm optimization (BPSO) algorithm. ML algorithm achieved high accuracy with lifestyle attributes. The ML algorithms were evaluated by deploying various measures like Accuracy, F1-Measures, Recall and Precision. The main aim of the paper is to develop a model by using ML techniques to help medical practitioners to make early prediction of diabetes.