Examining the Various Neural Network Algorithms Considering the Superiority of Mouth Brooding Fish in Data Classification

Abstract

Data classification, a crucial practice in information management, involves categorizing data based on its sensitivity to determine appropriate access levels and protection measures. This paper explores the utilization of novel algorithms, including mouth-brooding fish (MBF), alongside machine learning techniques, for the analysis of medical health data. The SVM exhibits suboptimal performance in the task of data categorization. Therefore, Adaboost may be considered a viable substitute for MBF due to its superior performance in terms of Fscore, accuracy, specificity, and sensitivity. The accuracy of MBF, which stands at about 95%, surpasses that of Adaboost by a significant margin of 77%. The F-score, accuracy, and specificity values obtained for MBF are exceptional when compared to the other chosen models, with values of 97.17%, 93.6%, and 96.5%, respectively. The proposed algorithm exhibits promising advancements in health data categorization, offering a potential breakthrough in data classification methodologies. Leveraging this innovative approach could facilitate more accurate and efficient management of sensitive medical data, thereby enhancing healthcare systems' capabilities for data protection and analysis. The main novelty of this study lies in the introduction and evaluation of the MBF algorithm for data classification within the medical domain. Unlike traditional algorithms, MBF draws inspiration from the collective behavior of mouth-brooding fish, offering a unique optimization strategy that enhances both exploration and exploitation of the solution space. This novel approach presents a promising avenue for advancing healthcare analytics and decision-making processes.