Predicting Stock Movements Using News Headlines And News Articles

Abstract

The main objective of the system is to analyse the future value of a certain stock of a particular company
using the sentiment analysis and to predict whether a particular stock will go up that is whether it will
increase or it will go down which means it will decrease on the basis of certain news headline, also detection
of fake news and OCR was implemented for providing the user as an option for entering the news headline
or a news article, the data we used was DJIA news headlines dataset and five different machine learning
algorithms were used – Random Forest classifier, Naïve Bayes, Decision Tree, Logistic Regression and
Support Vector Machine(SVM)