Preliminary Investigation in the use of Sentiment Analysis in Prediction of Stock Forecasting using Machine Learning

In recent years, many financial institutions have implemented machine learning applications to better predict stock prices. Computational methods such as regression or classification learners give financial analysts the ability to predict a stock's future price using historical data. More recen...

Full description

Saved in:
Bibliographic Details
Main Authors: Seals, Ethan, Price, Steven R.
Format: Conference Proceeding
Language:English
Subjects:
Computational modeling
Data integration
Feature extraction
Forecasting
Machine learning
Predictive models
Sentiment analysis
stock forecasting
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In recent years, many financial institutions have implemented machine learning applications to better predict stock prices. Computational methods such as regression or classification learners give financial analysts the ability to predict a stock's future price using historical data. More recently, sentiment analysis has been used to gather public opinion in hopes that it will provide meaningful information which would allow more accurate prediction models. In this work, we use both historical data and sentiment analysis scores to investigate whether sentiment analysis can improve performance of stock forecasting.
ISSN:1558-058X
DOI:10.1109/SoutheastCon44009.2020.9368258