| Abstrakt: |
The disease diagnosis in the medical field enhances better medical service to patients and also leads to a decrease in their mortality rate. The prediction of the survival rate of the patients purely depends on the accurate diagnosis of the diseases, but still, it is a major challenge to the physicians as well as to medical domains. Besides, several researches have experimented related to the prediction and classification of heart diseases, but they are ineffective in providing accurate results. In this research, the performance analysis of the optimal clustering algorithm-based real-world heart dataset is carried out with the developed clustering methods. Here, three developed methods, such as kernel-based exponential grey wolf optimisation (KEGWO), enhanced kernel-based exponential grey wolf optimisation (EKEGWO), and whale grey clustering (WGC) algorithm obtained better performance and provided accurate results about the diagnosis of diseases. Moreover, the performance analysis is done by considering the evaluation metrics like the Jaccard coefficient, F-measure, MSE and Rand coefficient. |
