Stock Price Movement Prediction Using Technical Analysis and Sentiment Analysis

Abstract

This study aims to predict stock price movement using combination of technical analysis and sentiment analysis. When conducting stock transactions, the traders consider not only market activities but also the sentiments expressed within information reported in media. We build the classifier to categorize the price quotes into one of three classes: "up", "down", and "constant". We conduct the experiment with several algorithms, i.e. Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Naïve Bayes. The results of our empirical study is that the highest accuracy achieved from the method combining features from historical data and online media sentiment, on 5 days trading window using the SVM algorithm.