| dc.description.abstract | Abstract: Diabetes is a deadly chronic
disease which affects entire body system
harmfully. Millions of people are affected by
this disease and a considerable number of
patients die every year because of its side
effects. A diabetic patient suffers from a high
level of blood sugar in the body.
Undiagnosed diabetes may cause the nerve
and kidney damage, heart and blood vessel
disease, slow healing of wounds, hearing
impairment and several skin diseases. Early
detection of diabetes is very essential to
have a healthy life. The recent development
of Machine Learning approaches solves this
kind of critical problems. The main objective
of this study is to present a Machine
Learning based solution (Artificial Neural
Network) to solve the above problem. And
also, the technologies and approaches used
in previous researches to predict diabetes
have been reviewed with their accuracy
levels. All the previous studies have used
“Pima Indian Diabetes Dataset” (PIDD) as
the dataset but this research is based on a
newly collected dataset. The overall
development process can be categorized
into four major development phases namely
data collection and preprocessing, statistical
analysis, development of machine learning
model and development of front-end.
Artificial Neural Network model has been
developed and deployed while the model
provides more than 92% accuracy on the
sample testing dataset. | en_US |
