A Study on Mathematics Modeling using Fuzzy Logic and Artificial Neural Network for Medical Decision Making System
Liton Chandra Voumik, Raj Karthik, Aditya Ramamoorthy, Anurag Dutta
Abstract
The conceptual formulation for diabetes & hepatitis in what seems like a fuzzy & neural network is investigated in this article. Because certain facts in a real-world situation cannot be recognized with precision, the metrics are used as approximate values. The membership functions and set of rules that make up the diabetes framework include fuzzy logic as their beginning premise. Fuzzy logic is just an augmentation of propositional logic which works with uncertainty and ambiguity in systems and offers a powerful mathematical tool to express data in a fashion that mimics naturally occurring human consciousness. ANN is a computational structure with biological inspiration made up of intricately linked adaptable simple processing units that can process information and convey information in a highly comparable manner. Investigators and academics from a diverse range of scientific and engineering disciplines were drawn to the emerging need for responsive intelligent machines because of the integration of Fuzzy Inference algorithms with Artificial Neural Networks. Fuzzy Systems, that can rationalize with imperfect information, are more adept at explaining their choices than artificial neural networks, which struggle to do so. Nevertheless, fuzzy systems are unable to learn the rules they employ to make decisions. Owing to these restrictions, intelligent hybrid systems have been developed to solve the issues using specific methodologies. The much more popular hybrid Neuro-Fuzzy approaches are discussed in this questionnaire survey, along with their benefits and drawbacks.