Predicting the stock market trend has always been challenging since its movement is affected by many factors. Here, we approach the future trend prediction problem as a machine learning classification problem by creating tomorrow_trend feature as our label to be predicted. Different features are given to help the machine learning model predict the label of a given day; whether it is an uptrend or downtrend, those features are technical indicators generated from the stock's price history. In addition, as financial news plays a vital role in changing the investor's behavior, the overall sentiment score on a given day is created from all news released on that day and added to the model as another feature. Three different machine learning models are tested in Spark (big-data computing platform), Logistic Regression, Random Forest, and Gradient Boosting Machine. Random Forest was the best performing model with a 63.58% test accuracy.