Forecasting Stock Value Based on Data from Social Media and Investment Instruments

This study aimed to predict stocks using different machine learning techniques with social media data and investment instrument data. Within the scope of the study, 236,764 tweets related to five different airline companies during the period October 2019 - February 2020, the stock value of those com...

Full description

Saved in:
Bibliographic Details
Main Authors: Ömer Faruk Uyrun, İbrahim Sabuncu
Format: Article
Language:English
Published: Istanbul University Press 2021-12-01
Series:Acta Infologica
Subjects:
Online Access:https://cdn.istanbul.edu.tr/file/JTA6CLJ8T5/5785252772664B2F805BF3DB51A33205
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study aimed to predict stocks using different machine learning techniques with social media data and investment instrument data. Within the scope of the study, 236,764 tweets related to five different airline companies during the period October 2019 - February 2020, the stock value of those companies, the daily data of the stock market, dollar rate and gold prices were discussed. Additionally, sentiment analysis was carried out in the analysis of the tweets. In the study, it was determined that the Gradient Boosted Trees algorithm was the prediction model with the lowest margin of error in stock prediction, and it was seen that the number of net positives (positive-negative) tweets about companies was one of the most influential factors in forecasting stock value. As a result of the study, it is thought that the Gradient Boosted Trees algorithm is effective in stock prediction compared to the other algorithms used in the study, and that Twitter data is a data source that can be used in forecasting stock value together with other investment data.
ISSN:2602-3563