![]() | Only 14 pages are availabe for public view |
Abstract It became obvious the role of machine learning algorithms in different fields which used to analyze data, values prediction, and classification. One of these fields is the finance field so there are different applications of machine learning in finance and one of these applications is the stock market value prediction and classification of its direction. A stock market is a place where individuals are buying and selling shares of publicly traded companies. A stock is a type of investment that represents an ownership share in a company and investors purchase stocks that they think will go up in value so machine learning algorithms are used to predict and classify the stock market value and its direction by gathering historical datasets. The purpose of predicting the stock market is to anticipate the price value and direction of stock as higher profits will investors can be made by higher accuracy prediction are gotten and here is one of the most challenging issues is predicting how the stock market will move. Neural Network has intended to mirror the elements of the human mind and this imitation of brain modeling permits the neural network to learn from experience without requiring human mediation and adjust appropriately to the circumstances. This thesis is going to predict the value of the close price and the direction of the stock market by applying two proposed models. The first proposed model is artificial neural network architecture and this is used to classify the stock market’s direction. The second proposed model is Long-short term memory neural network architecture and this is used to predict the close price value. Finally, the results of performance will be compared with other new published papers that were used the same dataset. |