The use of wind energy is on the rise in the United States and elsewhere.  However, predicting the production of power from wind is highly uncertain. In this paper, the effectiveness of three machine learning algorithms: ridge regression, polynomial regression, and artificial neural networks is compared on the predictive accuracy of the wind speed, which directly affects the wind power generation. Five-fold cross-validation technique is used to train and test the three algorithms using a range of hyperparameters. It is shown that the polynomial regression provides a better prediction than the other algorithms based on the root mean square error and R-squared metrics.