Prediction of stock value using pattern matching algorithm based on deep learning

Abstract

In this paper we began with finding ways to predict stock value flows of stock using deep learning. The purpose of this paper is to analyze the patterns in stock value and to analyze the relationship from stock values by deep running to predict what patterns will happen next stock value. In this paper we made the data by dividing the stock value information of the time series for a certain period of time and the pattern of stock value by analyzing these data. It is configured the model to be used for deep learning and learned the patterned time series information using the created model. And then it is predicted the next pattern of stock value. This paper focused machine learning. It is used of a time-series stock value information to predict the rise and fall of stock value. This paper is about how to analyze and how to predict. On the other hand, we can expect trend of stock value with high probability by analyzing pattern of current chart and anticipating pattern to follow. This is about what the deep-learning machine will analyze and predict for what. If we analysis the patterns used in this paper more clearly and concisely, and if more learning is carried out, we will be able to make clearer predictions with no noise for future trends. As interest in stock forecasts and machine learning develops fast, performance is expected to improve day by day.