TY - GEN
T1 - Machine Learning Algorithms for Early Prediction of Diabetes
T2 - 2022 International Conference on Electrical and Computing Technologies and Applications, ICECTA 2022
AU - Alzoubi, Rouaa
AU - Harous, Saad
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Diabetes is a chronic disease caused by increased blood glucose levels. Several physical and chemical tests can be used to diagnose this disease. Untreated and undiagnosed diabetes, on the other hand, can harm human organs such as the eye, heart, kidneys, and nerves and may even lead to death. As a result, early detection and analysis of diabetes can help reduce the death rate. Machine learning and deep learning models have been used recently in many medical fields, and their efficiency for the early diagnosis of different diseases has been noticed. This study aims to discuss the different state-of-the-art algorithms that researchers have implemented for the early prediction of diabetes. The work focuses on highlighting different techniques used in the literature and the effectiveness of those techniques, which can help in knowing the current limitations of the work and making more improvements to it. As a result, our research showed that the random forest and KNN algorithms outperformed other algorithms in the literature with an accuracy of 98% in the early prediction of diabetes.
AB - Diabetes is a chronic disease caused by increased blood glucose levels. Several physical and chemical tests can be used to diagnose this disease. Untreated and undiagnosed diabetes, on the other hand, can harm human organs such as the eye, heart, kidneys, and nerves and may even lead to death. As a result, early detection and analysis of diabetes can help reduce the death rate. Machine learning and deep learning models have been used recently in many medical fields, and their efficiency for the early diagnosis of different diseases has been noticed. This study aims to discuss the different state-of-the-art algorithms that researchers have implemented for the early prediction of diabetes. The work focuses on highlighting different techniques used in the literature and the effectiveness of those techniques, which can help in knowing the current limitations of the work and making more improvements to it. As a result, our research showed that the random forest and KNN algorithms outperformed other algorithms in the literature with an accuracy of 98% in the early prediction of diabetes.
KW - Diabetes
KW - Diabetes mellitus
KW - Disorder
KW - Machine learning
UR - http://www.scopus.com/inward/record.url?scp=85146368811&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85146368811&partnerID=8YFLogxK
U2 - 10.1109/ICECTA57148.2022.9990240
DO - 10.1109/ICECTA57148.2022.9990240
M3 - Conference contribution
AN - SCOPUS:85146368811
T3 - 2022 International Conference on Electrical and Computing Technologies and Applications, ICECTA 2022
SP - 401
EP - 405
BT - 2022 International Conference on Electrical and Computing Technologies and Applications, ICECTA 2022
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 23 November 2022 through 25 November 2022
ER -
