Stock price prediction with machine learning is an oft-studied area where numerous unsolved problems still abound owing to the high complexity and volatility that technical-factors and sentiment-analysis models are trying to capture. Nearly all areas of machine learning (ML) have been tested as solutions to generate a truly accurate predictive model. The accuracy of most models hovers around 50%, highlighting the need for further increases in precision, data handling, forecasting, and ultimately prediction. In this paper we present the result of our work on highfrequency (every fifteen minutes) stock-price prediction using technical data with a number of exogenous variables. These variables are carefully chosen to reflect the conventional wisdom in a traditional stock analysis on historical trend, general stock market condition, and interest rate movement. Several simple machine learning (ML) algorithms were developed to test the premise that with the appropriate variables, even a simple ML model could produce reasonable prediction of stock prices. Therefore, the originality of our approach is a rational selection of relevant and useful features and also on-the-fly model retraining taking advantage of the human time scale of inference (price prediction) and moderate size of the models. Moreover we do not mix any trading strategy with our stock-price prediction experiments, to ensure that conclusions are not context-dependent. Systems that integrate and test sentiment and technical analysis are considered the best candidates for an eventual generalized trading algorithm that can be applied to any stock, future, or traded commodity. However, much work remains to be done in applying natural language processing and the choice of text sources to find the most effective mixture of sentiment and technical analysis. Work on this area will be included in the next phase of our research project and here we have summarized some of the most relevant existing works in this direction.