Artificial Intelligence with uncertainty
Lincy Jacquline M; Sudha N
Abstract
Problem Statement: Chronic nephritic sickness is another name for chronic kidney disease (CKD). Numerous complications, such as elevated blood levels, anemia, weak bones, and nerve damage, constitute a problem. It is usually possible to prevent chronic uropathy from getting worse by early identification ...
Read More
Problem Statement: Chronic nephritic sickness is another name for chronic kidney disease (CKD). Numerous complications, such as elevated blood levels, anemia, weak bones, and nerve damage, constitute a problem. It is usually possible to prevent chronic uropathy from getting worse by early identification and treatment. Methodology: To circumvent these problems, current research has presented the fruit fly optimization algorithm (FFOA) and effective multi-kernel support vector machine (MKSVM) for illness classification. Finding best features from a collection is usually done using FFOA. Main findings/Contributions: MKSVM categorizes medical data using chosen dataset criteria. The accuracy of classifier will be impacted by any range variations in data obtained for this study. MKSVM continues to yield more incorrectly classified findings. To resolve those problems introduces a pre-processing step based on min max normalization to normalize scale of input CKD data values. Then significant features will be selected utilizing Improved FFOA (IFFOA). The selected features will be clustered using Weighted Fuzzy C means clustering (WFCM) to predict the class label of the data sample to reduce the misclassification results. Finally, CKD classification will be performed using the Enhanced Adaptive Neuro Fuzzy Inference System (EANFIS) as normal or abnormal. Conclusions: The suggested strategy efficacy is demonstrated by findings in fields of recall, accuracy, precision, and f-measure.
