Online ISSN: 2515-8260

Author : Ahamed, B. Shamreen

LGBM Classifier based Technique for Predicting Type-2 Diabetes

B. Shamreen Ahamed; Dr. Meenakshi Sumeet Arya

European Journal of Molecular & Clinical Medicine, 2021, Volume 8, Issue 3, Pages 454-467

In today’s world, Diabetes Mellitus is a disease, that is considered to be an extensive noncommunicable disease which has a great effect our day to day living. In the 21st century, changes in natural life style and labor culture are some of the main reasons for India to have 62 million diabetic cases as of today. Analytical Computational Techniques can be applied on clinical immense data, the enormous quantity of data produced in the healthcare schemes, there is a option to form medicinal intelligence which will initiative medical forecast andpredicting in future. By advancing medical intelligence and with the help of development model, prediction and detection of diabetes disease can be done. With the increase in complexity to the problems, the accuracy percentage also varies. LGBM - Light Gradient Boosting Algorithm is one such algorithm that can be used as it depends on decision tree algorithms and it can be used in predicting the accuracy to attain the desired results. With the
existing PIMA Indian Dataset the accuracy is calculated as 95.20% using LGBM Algorithm . Therefore by using the LGBM classifiers, we can develop a data model for diabetes detection and prediction.