Efficient tagging of questions on online platforms like Stack Overflow is crucial for enhancing search functionality and improving user experience. This study investigates the effectiveness of various machine learning algorithms for autonomous tagging of Stack Overflow questions. We employ a diverse range of models, including XGBoost, LinearSVC, Stochastic Gradient Descent Classifier, Logistic Regression, Perceptron, and PassiveAggressiveClassifier, and evaluate their performance using a comprehensive set of metrics, including Jaccard score, Hamming loss, and F1 score. Our results show that XGBoost and LinearSVC perform better than the other models, indicating their capacity to minimize misclassifications and predict many relevant tags with accuracy. Both models appear to be well-suited for the task of autonomous tagging, albeit there is a slight performance difference between them. Our research advances autonomous tagging methods by comparing different machine learning algorithms and determining which models perform best in this context.