Genetic Algorithm-Driven Weight Adjustment in Ensembles for Advanced Thyroid Classification

The thyroid is the tiniest gland on the front of the neck, shaped like a butterfly. It generates hormones that regulate the amount of energy the body uses. This research explores the use of machine learning techniques, ensemble learning strategies, and optimization algorithms for ensemble weights in response to the vital importance of thyroid diagnosis. To accurately identify thyroid disorders, a wide range of machine learning methods are used, including logistic regression, Support Vector Classifier (SVC), MLP Classifier, KNN Classifier, Decision Tree Classifier, and Random Forest Classifier. In addition, the application of ensemble learning techniques demonstrates the enhanced performance that results from combining these models. This research provides insights into the effectiveness of Genetic EnsembleWeight Adjuster in changing the ensemble weights, hence leading to improved thyroid classification accuracy. The Random Forest Classifier proved to be effective in identifying thyroid abnormalities in this investigation, as seen by its best accuracy of 98.9%. A remarkable accuracy of 99.9% was attained by adjusting the weight of the Ensemble Approach using the Genetic EnsembleWeight Adjuster and based on the experimental result it gives better accuracy compared to the Ensemble Approach without adjusting weights and other classifiers.