For the design and implementation of CDSS, computation time and prognostic accuracy are very important. To analyze the large collection of a dataset for detecting and diagnosis disease ML techniques are used. According to the reports of World Health Organizations, HD is a major cause of death and killer in urban and rural areas or worldwide. The main reason for this is a shortage of doctors and delay in the diagnosis. In this research work, heart disease is a diagnosis by the data mining techniques and used the clinical parameters of patients for early stages diagnosis. The intend of this learning to develop a representation that relies on the prediction method for coronary heart disease. This proposed work used the approach of self-diagnosis Algorithm, Fuzzy Artificial neural network, and NCA & PCA and imputation methods. By the use of this technique computation time for prediction of Coronary HD can be reduced. For the implementation of this the two datasets are using such as Cleveland and Statlog datasets that is collected from the UCI kaggle the ML repository. The datasets for the disease prediction measure are used to accurately calculate the difference between variables and to determine whether they are correlated or not. For this classification model, the performance measure is calculated in requisites of their accuracy, precision, recall, and specificity. This approach is evaluated on the heart disease datasets for improving the accuracy performance results obtained. The outcome for KNN+SDA+NCA+FuzzyANN for Cleveland dataset accuracy achieved 98.56 %.and for Statlog dataset 98.66 %..
CITATION STYLE
Aggarwal, R., & Kumar, S. (2022). Classification Model for Meticulous Presaging of Heart Disease Detection through SDA and NCA using Machine learning:CMSDANCA. International Journal on Recent and Innovation Trends in Computing and Communication, 10(1), 217–224. https://doi.org/10.17762/ijritcc.v10i1s.5827
Mendeley helps you to discover research relevant for your work.