Systematic Mapping Study on Applications of Deep Learning in Stock Market Prediction

Stock price forecasting is a challenging and complex problem with significant earnings potential for investors and institutes. Thereby, investors, economists and scientists, have studied and developed different approaches to predict stock trend or price. Deep learning methods have shown promising results in different areas such as audiovisual recognition and natural language processing. In the same way, researchers are exploring the effectiveness of the application of deep learning to the problem of stock price prediction. This study aims at reviewing and mapping the works on deep learning applications to stock price prediction. By collecting, analyzing and classifying the existing literature, we explore and give an overview of the latest progress in the studied topic and identify possible future research directions.