A Novel Framework for Classification of Diabetes Diseases Patients Using Random Forest as Feature Selection

Authors

  • Zeenat Bashir
  • Dr. Hamid Ghous

DOI:

https://doi.org/10.5281/zenodo.5148271

Keywords:

Diabetes, decision tree (DT), neural network (NN), random forest (RF), support vector machine (SVM)

Abstract

Technological advancements are increasing day by day in biomedical field. A huge amount of data has been collected from different resources and used in the biomedical field. Data mining is a process of extracting the hidden patterns from large datasets to gain useful information for users. However, data mining trend in healthcare applications is increasing, and it is playing an important role in the medical field for diagnosing and predicting diseases at early stages. Data mining techniques such as classification, clustering, association, regression, and summarization have been previously used for diagnosing and predicting diseases. Diabetes is a common and chronic disease which causes an increase in blood sugar. Many complexities occur if diabetes remains unidentified and untreated. The present study aims are to implement Random Forest (RF) as a feature selection method and some classification method such as Support Vector Machine (SVM), Decision Tree (DT) and Neural Network on two diabetes dataset for the early diagnosis of diabetes. The proposed results show that the Support Vector Machine provides higher accuracy for the prediction of diabetes disease. It will be very effective and efficient for everyone.

Downloads

Download data is not yet available.

Downloads

Published

2020-10-06

How to Cite

Zeenat Bashir, & Dr. Hamid Ghous. (2020). A Novel Framework for Classification of Diabetes Diseases Patients Using Random Forest as Feature Selection. LC International Journal of STEM (ISSN: 2708-7123), 1(3), 32-49. https://doi.org/10.5281/zenodo.5148271