.png){kind=icon}
返回
RG-SVM: Recursive gaussian support vector machine based feature selection algorithm for liver disease classification
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-023-17825-1.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Health is an essential concern for everyone, so it is necessary to facilitate medical services that are easily accessible to everyone. The primary goal of this work is to predict liver diseases using a machine-learning strategy that makes use of feature selection and classification techniques. This work proposes the recursive Gaussian support vector machine-based feature selection (RG-SVM) algorithm. It uses the Gaussian kernel of support vector machine and recursive feature selection algorithm for the prediction of liver disease. The proposed RG-SVM algorithm has been evaluated on the Indian liver patient records dataset. Various classification algorithms such as logistic regression, decision tree, k-nearest neighbour, and Naive Bayes are implemented and compared in order to assess the accuracy, confusion matrix and area under curve. The proposed RG-SVM has been compared with other existing algorithms such as logistic regression (LR), decision tree (DT), k-nearest neighbour (KNN), Naive Bayes (NB), and proposed RG-SVM algorithms. The algorithms LR, DT, KNN, NB, and proposed RG-SVM have accuracy values of 73, 80, 81, 54, and 93%, respectively. It clearly shows that the proposed RG-SVM with the support of a recursive feature selection algorithm, outperformed other existing algorithms with an improved accuracy of 14 - 39% 12- 20% of reduced MSE error over other compared algorithms. Similarly, the sensitivity and specificity of RG-SVM algorithm produced 5-26% and 34-72% improved results over the existing algorithms. The results of the proposed algorithm will be useful for physicians to make better decisions for liver disease patients.
Keyword:
Recursive Feature selection
Feature ranking, RG-SVM, decision tree
Liver disease Prediction
期刊
IF:
3
论文数:
2.0W
被引数:
3.2W
机构
引用论文
.png){kind=icon}
分享收藏分享收藏分享收藏
查看更多内容