A first attempt to develop a diabetes prediction method based on different global datasets

Diabetes is a chronicle disease which increases with time and cause various other complications if not treated in early stage, it is also a reason of early death in diabetes patients due to other complication. There are so many peoples all over the world who are suffering from diabetes and still undiagnosed. Many researchers have developed computational methods for diabetes diagnosis but those systems are not widely applicable and reliable because these methods are tested and trained on single dataset. Considering the global nature of diabetes there is need of a method which is tested trained and validated on all type of datasets or on all the available dataset. No such method have been still developed which is tested, trained and validated on different dataset. In the present study a method is developed using combined datasets using machine learning technique. This system is more reliable because of trained, tested and validated on combine dataset.