Review of Predicting Stock Prices Based on Machine Learning and Stock Learning

Abstract

Predicting stocks is to help investors make decisions and help them minimize risks as much as possible, thereby obtaining investment returns. If we can accurately predict the rise and fall of stock market prices, we can buy or sell stocks at higher prices in the market, thereby obtaining higher profits. This article is based on machine learning and deep learning algorithms such as support vector machine (SVM), principal component analysis (PCA), random forest (RF), and recurrent neural network (RNN), long short-term memory network (LSTM), and convolution neural network (CNN). The conclusion is that machine learning requires manual intervention, is faster to train, more interpretable, and usually performs better on data sets with simpler structures and smaller sizes. Deep learning has lower requirements for feature extraction and discovery, has stronger expression and fitting capabilities, and can better capture the correlation between data in larger and more complex data sets.