Predicting Short-Term Exchange Rates for Automatic Purchasing Using News Article Data

Recent years, increasing online securities companies has led to a massive increase in the number of investing in stock and foreign exchange markets by individual investors. For individual investors, text data such as news articles and websites is a familiar and easy source of information. In addition, machine learning techniques for stock market prediction have attracted much attention by using text data as well as numerical data. Due to the small number of studies on exchange rates, it also increases the demand for supporting exchange investment. This paper introduces a novel method to predict short-term price movements of exchange using machine learning techniques by utilizing online news articles. In this work, our proposed method can detect the relationships between news articles and exchange rate fluctuations. Furthermore, we apply foreign exchange forecasting to automatic purchasing systems, and we verify the usefulness of the proposed methods support for exchange investment.