Predictive Analysis of Medical Data using a Hybrid Machine Learning Technique

Diabetes is a chronic disease, which affects millions of people around the world. It causes serious damage to human health and even sometimes human lives. Various research is going on in this area to reduce its negative impact. In this paper, we propose a hybrid machine learning technique to classify people into patient and healthy classes, respectively. The patient class represents people with diabetes and the healthy class represents people with no diabetes. We evaluate the proposed model against various performance measures and found that it performs better than previously used techniques. We use Pima Indian Diabetes dataset, which has 768 instances and 9 attributes. Out of 768 instances, 268 are affected and 500 are not affected by diabetes.