Machine Learning Application for Stock Market Price Prediction

ABSTRACT The development of a vibrant application for analyzing and predicting stock market prices is a basic tool aimed at increasing the rate of investors' interest in stock markets. This paper explains the development and implementation of a stock price prediction application using machine learning algorithm and object oriented approach of software system development. The algorithm was used in training a set of market data collected for the period of one thousand, two hundred and three days. The implementation was done with Java programming language and Neuroph software. From the experiments conducted and observed results from the application indicates a better predictive accuracy and high minimum error. When tested with Nestle Nigerian Plc. recorded a mean squared error (MSE) and regression (R) values of 61876e-6 and 0.99975 respectively, Guinness Nigerian Plc. recorded a mean squared error (MSE) and regression (R) values of 9.95839e-7 and 0.99853 respectively and Total Nigerian Plc. recorded a mean squared error (MSE) and regression (R) values of 8.03493e-6 and 0.992193, respectively. (Keywords: machine learning, supervised learning, regression value, mean square error