Stock price prediction using decision tree classifier and LSTM network

Abstract

Nowadays, stock price prediction has become a popular research topic, many researchers try to predict stock prices in various ways. However, there are many different tools, but not all of them have good performance, so it is necessary for researchers to evaluate and compare different tools. In this paper, to achieve the goal of predicting stock price precisely, the main approach chosen is building deep learning models and use them to make predictions. Two methods, decision tree and long short-term memory (LSTM) neural network, are used in this study. In the model using the decision tree classifier, the daily state of the stock is divided into two types: the rise and fall of the stock price. The task of the model is to make predictions about daily stock prices and classify them. The other model uses the LSTM network, which is used to make accurate closing price predictions. In the end, the performance of the two models is assessed for further work.