Stock market prediction is an attempt of determining the future value of a stock traded on a stock exchange. This project focuses on classification problems, predicting the next-second price movement, and acting upon the insights generated from our models. We implemented multiple machine learning algorithms including logistic regression, support vector machines (SVM), Long- Short Term Memory (LSTM), and Convolutional Neural Networks (CNN) to determine the trading action in the next minute. Using the predicted results from our models to generate the portfolio value over time, the support vector machine with a polynomial kernel performs the best among all of our models.