A Hybrid Model of Integrating Sentiment Analysis and Key Market Indicators for IPO Listing Trend Prediction

Abstract

Initial Public Offerings (IPOs) provide great opportunities for companies to grow and expand, and they allow investors to invest their money wisely to get a decent Return on Investment (ROI) in the short term. Nevertheless, the intricate nature of the stock market is susceptible to several influences such as a company's financial statement, governmental regulations, and public sentiment, which hinders the attainment of a satisfactory ROI. This study aims to address this challenge by developing a model that combines public sentiment analysis and machine learning approaches to optimize the ROI for IPO trends. The study gives a novel approach that uses multiple different features associated with IPOs like the public opinion, grey market price (GMP), issue prices, lot size etc. and leverages the use of various machine learning techniques like Random Forest, Decision Tree, Naive Bayes, and K-Nearest Neighbour (KNN) to make well-informed investment recommendations. The testing results demonstrate that the Decision Tree method surpasses the other algorithms, with an accuracy rate of 82.3%. This discovery emphasizes the effectiveness of our method in forecasting the success of IPOs by utilizing a combination of sentiment analysis and crucial financial indicators.

Keywords

• sentiment analysis
• ipo
• machine learning algorithm
• gmp
• decision tree
• issue prices