An efficiency of DCKSVM and HRBFNN techniques for diabetic prediction

Data and knowledge discovery are essential in the healthcare industry for disease diagnosis and prediction. Patient records and disease-related information are data in the health industry. The DCKSVM and HRBFNN approaches are utilized to predict the result. The aim is for evaluating the efficiency o...

Full description

Saved in:
Bibliographic Details
Main Authors: Sivaraman, M., Sumitha, J.
Format: Conference Proceeding
Language:English
Subjects:
Algorithms
Machine learning
Neural networks
Predictions
Radial basis function
Support vector machines
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data and knowledge discovery are essential in the healthcare industry for disease diagnosis and prediction. Patient records and disease-related information are data in the health industry. The DCKSVM and HRBFNN approaches are utilized to predict the result. The aim is for evaluating the efficiency of the DCKSVM-Divide and Conquer Kernal Support Vector Machine learning and RBFNN-Radial Basis Function Neural Network algorithm in detecting diabetes in the general population. Diabetic prediction can be predicted by many existing algorithms, but machine learning algorithms for used this research. In this research, the KNN algorithm is used as a preprocessing tool to fulfill the missing attribute value in the dataset. The HRBFNN approach gives good accuracy. The result shows that the RBFNN algorithm give good accuracy than SVM in predicting diabetics.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0162883