Presaging The Signs Of Diabetes Using Machine Learning Algorithms

Diabetes is an acute illness that occurs mainly when the glucose levels in the blood go up due to an insufficient amount of insulin available to the body. Additionally, a large number of deaths across the globe are attributed to diabetes every year. Machine Learning techniques have evolved exponentially and have been widely adopted in various medical applications. These techniques help in deducing the hidden patterns in the data and further, help in disease predictions based on these deduced patterns. They have also fetched good results in predicting diseases like diabetes and have proven useful in the early diagnosis of the disease. This paper focuses on the important steps like finding the correlation of the different health parameters essential for Diabetes prediction like Glucose, Diabetes Pedigree Function, BMI, and others. Further, a comparison of popular Machine learning approaches like Logistic Regression, Decision Tree, K-Neighbours, and Naive Bayes Classifiers have been made to determine the best technique based on the performance metrics.