Sentiment Analysis based Stock Price Prediction using Machine Learning

This paper endeavors to contribute to the everevolving landscape of stock market forecasting by leveraging Sentiment Analysis (SA) on Twitter data and using it to predict the stock price using the following Machine Learning (ML) regression models: K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Linear Regression and Random Forest (RF). Natural Language Processing (NLP) tools such as VADER and TextBlob have been used to extract market sentiment information. The dataset includes data of ICICI bank along with 73762 tweets related to the company during the period of 1-Jan-2018 to 15-Jan-2023. During the evaluation, out of the four ML models, Linear Regression was found to be the most accurate with an $R^{2}$ value of 0.8741 followed by values of $0.6783,0.0120,-0.2820$ for SVM, RF and KNN respectively with the results of Vader and $0.8741,0.6860$, 0.1133 and -0.0304 respectively when considering TextBlob.