An Innovative Artificial Intelligence-Based Extreme Learning Machine Based on Random Forest Classifier for Diagnosed Diabetes Mellitus

Saputra, Dimas Chaerul Ekty and Muryadi, Elvaro Islami and Phann, Raskmey and Futri, Irianna and Lismawati, Lismawati (2024) An Innovative Artificial Intelligence-Based Extreme Learning Machine Based on Random Forest Classifier for Diagnosed Diabetes Mellitus. Jurnal Ilmiah Teknik Elektro Komputer dan Informatika (JITEKI), 10 (1). pp. 173-187.

Text 12-An Innovative Artificial Intelligence-Based Extreme Learning Machine Based on Random Forest Classifier for Diagnosed Diabetes Mellitus.pdf Download (753kB)

Abstract

Since 2014, the World Health Organization has accumulated data indicating that 8.5% of 18-year-olds and older have been diagnosed with diabetes. In 2019, diabetes caused the lives of 1.5 million people worldwide, with those under the age of 70 accounting for 48% of all diabetes-related deaths. It is estimated that diabetes causes an additional 460,000 deaths each year due to renal failure and that hyperglycemia contributes to about 20% of all cardiovascular disease-related deaths. Diabetes may have contributed to a 3% rise in the age-adjusted death rate between the years 2000 and 2019. In recent years, the fatality rate attributable to diabetes has increased by 13% in low- and middle-income countries. Statistics collected by the World Health Organization indicate that the number of persons diagnosed with diabetes has increased from 108 million in 1980 to 422 million in 2014. The objective of this study is to construct a model capable of diagnosing persons with diabetes reliably, correctly, and consistently. This research used secondary data offered by Kaggle. The original data came from the National Institute of Diabetes and Digestive and Kidney Diseases. Each of the up to 768 data points consists of nine characteristics and two outputs, such as diabetes and non-diabetes in the provided example. In this study, a single algorithm is constructed by integrating two separate algorithms. Random forest algorithms, which are based on machine learning, and extreme learning machines, which are based on deep learning, have generated extraordinarily accurate results. When the confusion matrix is used, 98.05% accuracy is attained. Therefore, it is feasible to conclude that the suggested method was successful in completing an adequate analysis and classifying the data.

Item Type:	Artikel Umum
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisi / Prodi:	Faculty of Industrial Technology (Fakultas Teknologi Industri) > S1-Electrical Engineering (S1-Teknik Elektro)
Depositing User:	M.Eng. Alfian Ma'arif
Date Deposited:	21 May 2024 03:57
Last Modified:	21 May 2024 03:57
URI:	http://eprints.uad.ac.id/id/eprint/63561

Actions (login required)

View Item