Title: Multilayer perceptron sustainable boosting algorithm for thyroid classification systems
Authors: V. Brindha; A. Muthukumaravel
Addresses: Department of Computer Applications, Bharath Institute of Higher Education and Research, Chennai, 600073, Tamil Nadu, India ' Department of Computer Applications, Bharath Institute of Higher Education and Research, Chennai, 600073, Tamil Nadu, India
Abstract: Medical professionals can use sophisticated technologies like computer-aided diagnosis systems to help them make the best decisions possible by using data mining to analyse clinical data. When the thyroid gland cannot produce adequate thyroid hormones, a potentially fatal condition known as thyroid disease develops. This research aims to predict thyroid disease from the given dataset. This study suggests using the multilayer perceptron (MLP) technique to identify thyroid disease using straightforward criteria. In order to get the most out of the algorithms, this study evaluated the suggested strategy with the XGBoost classification technique using the UCI thyroid disease dataset. The obtained findings demonstrate the MLP model's capacity to accurately and precisely predict the proper type of thyroid problem diagnosis. According to the results, the MLP algorithm surpasses the XGBoost method with accuracy and precision values of 93% and 89%, respectively. The tool used for analysing the work is Jupyter Notebook, and the language used is Python.
Keywords: thyroid disorder; XGBoost; MLP; multilayer perceptron; endocrine; hypothyroidism; hormone; diagnose; antithyroid; radial basis function; back-propagation; bagging; ensemble.
DOI: 10.1504/IJSSE.2024.138344
International Journal of System of Systems Engineering, 2024 Vol.14 No.3, pp.278 - 290
Received: 13 Jan 2023
Accepted: 27 Feb 2023
Published online: 01 May 2024 *