Enhancing the Performance of Classifiers in Detecting Abnormalities in Medical Data using Nature Inspired Optimization Techniques

A significant variety of nature-inspired techniques are used in the field of Machine Learning (ML) to handle a wide range of problems. The term “evolutionary computation” refers to a group of global optimization algorithms inspired by biological evolution. They're a type of population-based trial-and-error problem solver that includes a metaheuristic or stochastic optimization function. The aim of this work is to create a model that can effectively detect diabetes from the given dataset using nature inspired algorithms. This work implants the usage of nature inspired optimization techniques for predicting diabetes. The nature inspired algorithms such as Bat Algorithm, Hybrid Bat Algorithm, Grey Wolf optimization algorithm, and Firefly algorithm were mostly used for numerical data and also used for improving the accuracy and other performance metrics. Seven classification algorithms were used in detection of diabetes and the Nature inspired algorithms are used to tune the hyperparameters of the classifiers in order to get an enhanced accuracy. Majority of the used classifiers' performance was enhanced with the use of nature inspired algorithms. Firefly algorithm and hybrid bat algorithm outperforms the other algorithms with the highest accuracy of 77.2% in the case of random forest classifier.