Prices on the stock market vary frequently as a result of many economic, political, and social reasons. It is a fluid and challenging environment. Investors look for efficient tools to enhance their investing strategy and make well- informed judgements. This study investigates how stock market fluctuations are predicted using machine learning approaches and how well they can identify complex patterns. Various algorithms, including Logistic Regression, K-Nearest Neighbours, Support Vector Machines, Random Forest, Decision Tree, Naive Bayes, Long Short-Term Memory, Multilayer Perceptron, and XG Boost are evaluated and compared based on their accuracy, precision, recall, and F1 score. The Multilayer Perceptron emerges as the most accurate predictor, showcasing its ability to handle complex relationships and learn from historical data. The results of this study provide insightful information for investors who want to use predictions from machine learning in their decision- making.