UTILIZING A FUSION OF MACHINE LEARNING TECHNIQUES FOR DIABETES MELLITUS SUBTYPES CLASSIFICATION AND IDENTIFICATION
Keywords:
Classification, Prediction, Diabetes subtypes, Support Vector Machine, Random Forest, MisdiagnosisAbstract
Diabetes Mellitus (DM) is one of the most common health challenges in the world we live in today. It is a deadly disease which prevents the body from making enough insulin. Diabetes Type1 and Type2 are the two major types, which have some similarity in symptoms. Identifying Diabetic Patients with respect to type plays a very significant role in the management process. Misdiagnosis of these types leads to serious impediments. Research shows that the overlapping nature of features contributed to the difficulty in identifying the types and the classification into sub-types. This is still an area of concern (Hassan, et al, 2020; Albahli, 2020). In this research, we proposed a method of Support Vector Machine (SVM) and Random Forest Tree (RFT) for the classification of Diabetes sub-types. To reduce the dimensions of the feature set, the Principal Component Analysis (PCA) and Logistic Regression (LR) were used. For effective research, data is sourced from the Center for Endocrinology and Diabetes-Al-Kindy Teaching Hospital and Medical City Hospital's public laboratory Dataset to ensure wide coverage. The dataset consists of 834 patient records with eight features and an output column labelled "Type I" or "Type II." This study conducted the experiment using Python, and the results show that the hybrid model outperformed the other prediction methods.
Published
How to Cite
Issue
Section
FUDMA Journal of Sciences
How to Cite
Most read articles by the same author(s)
- Malik Adeiza Rufai, Ahmad Shehu Muhammad, Suleiman Garba, Lovingkindness Audu, MACHINE LEARNING MODEL FOR BREAST CANCER DETECTION , FUDMA JOURNAL OF SCIENCES: Vol. 4 No. 1 (2020): FUDMA Journal of Sciences - Vol. 4 No. 1
- Joseph Ndagatsa Mamman, Muhammad Bashir Abdullahi, John Kolo Alhassan, Opeyemi Aderike Abisoye, Oluwaseun Adeniyi Ojerinde, A BLOCKCHAIN-DRIVEN VAT COMPLIANCE MODEL USING HYPERLEDGER FABRIC AND MONTE CARLO SIMULATIONS , FUDMA JOURNAL OF SCIENCES: Vol. 8 No. 6 (2024): FUDMA Journal of Sciences - Vol. 8 No. 6 (Special Issue)