<![CDATA[Abstract. In this study, we presents a comprehensive analysis of gender identification methods utilising eight distinct classification models: K-Nearest Neighbors (KNN), Naive Bayes, Decision Tree, Random Forest, Logistic Regression, XGBoost, Support Vector Machine (SVM), and Neural Network. Gender identification is a critical task with significant applications in marketing, social analysis, and security systems, necessitating the exploration of various methodologies to achieve optimal performance. The dataset employed in this research underwent normalisation using the Min-Max scaling technique, which enhances the performance of classification models by ensuring that all features contribute equally, particularly when the data exhibits varying ranges of values. The results reveal that the K-Nearest Neighbors (KNN) model significantly outperformed the other models, achieving an impressive accuracy of 0.9758 with a support of 951, underscoring the effectiveness of the KNN algorithm in gender identification tasks and establishing it as a reliable choice for applications requiring high accuracy. Furthermore, the study emphasises the critical importance of selecting appropriate models in machine learning tasks and the substantial impact of data normalisation on model performance. Overall, this research provides valuable insights into the KNN algorithm, demonstrating its ease of implementation and exceptional effectiveness in achieving high precision in gender identification tasks, with implications for future research and practical applications across various fields. Keywords : classification models; data normalisation; gender identification; K-Nearest Neighbours; machine learning.]]>
Copyrights © 2024
