Crude oil plays a vital role in the global economy, making accurate price predictions essential for companies and investors in the oil industry. However, traditional methods of forecasting crude oil prices have limitations, prompting the exploration of machine learning algorithms as a potential solution. Machine Learning algorithms have shown great promise in predicting crude oil prices by effectively for large datasets and uncover complex. In the paper, historical data is use to train and test a neural network model, consisting of different features such as crude oil production levels, geopolitical events, and economic indicators to predict the daily closing price of crude oil. The paper primarily focuses on crude oil price forecasting using different machine and deep learning algorithms, including Linear Regression, XGBoost, Support Vector Regression (SVR), and CNN-LSTM. Overall, it highlights the potential benefits of employing machine-learning algorithms in crude oil price forecasting. These algorithms possess the ability to uncover intricate relationships and patterns and can prove to be a gain to decision makers in the oil industry with more accurate predictions, which will help to do strategies and investments more wisely.