Stock Price Prediction Using LSTM and News Sentiment Analysis

Abstract

The stock market is a place of interest both for the investors and researchers. Its movement is governed by a lot of factors all over the world. The objective of this research is to help in the prediction of stock prices with better accuracy. In this regard, an artificial intelligence technique is used. The proposed model has two different parts, one to analyze the pattern of movement of stock using LSTM and then predict on it. It gives an average RMSE score of 0.33. The other part is to deal with sudden stock movement due to news using sentiment analysis which can predict the stock movement direction. It gives an accuracy of 84.9%.