Open AccessDiagnosis of Diabetes Using Naïve Bayes Classifier Method
Author(s) -
Tasya Ardhian Nisaa,
Shavira Maya Ningrum,
Berlianda Adha Haque
Publication year - 2021
Publication title -
international journal of data science, engineering, and analytics
Language(s) - English
Resource type - Journals
eISSN - 2807-1689
pISSN - 2798-9208
DOI - 10.33005/ijdasea.v1i1.4
Subject(s) - naive bayes classifier , polyuria , diabetes mellitus , computer science , classifier (uml) , artificial intelligence , machine learning , python (programming language) , medicine , endocrinology , support vector machine , operating system
Not a few people suffer from diabetes, diabetes is usually caused by genetic inheritance from parentsand grandparents. Not only from heredity but many criteria or characteristics can determine a person hasdiabetes. This research was conducted by looking for a dataset on Kaggle that contains criteria for someonediagnosed or undiagnosed with diabetes such as age, gender, weakness, polyuria, polydipsia, and others.Furthermore, from these criteria, predictions are calculated using the Naive Bayes classification methodwhere this method is one of the data mining techniques. This prediction calculation uses the Pythonprogramming language. From these criteria, each criterion is grouped with similarities and the results ofthe program that have been made can diagnose someone with diabetes. The prediction calculations that havebeen carried out have resulted in 90% accuracy, 93% precision, 89% recall, 92% specificity, and 91%F1-Score.