Feature Extrusion and Categorization of Disease by Hybrid Neuro‐Fuzzy Computing
Manideep Yenugula, K.S. Chakradhar, Makhan Kumbhkar, D. Victorseelan, Rupinder Karur · 2025
Medical illness categorization using machine learning algorithms encounters difficulties due to insufficient, unclear, and erroneous data. The performance of classification models is affected by the availability of data. The research in this article classifies illnesses using medical data using a model called Linguistic Neuro-Fuzzy Feature Extraction (LNF-FE). To deal with uncertainty, the first model uses linguistic fuzzification to derive membership values. While increasing membership values may not have a major effect on the system, it will increase the number of aspects, which means more time is needed for training. To address this issue, the Neuro-Fuzzy (NF) model employs a combination of Feature Extraction (FE) algorithms to determine and extract the most valuable properties for the network. The artificial neural network (ANN) method is used for categorization with these decreased features. They compare the proposed model's performance to that of existing models and test and verify it using eight benchmark datasets. Statistical methods like Friedman and Holm-Bonferroni were used to verify the accuracy of the findings. The results of these experiments demonstrate that, when applied to real-world issues, the proposed approach performs better than competing methods.