Enhancing stock market predictions for classifying unlabelled celebrities' twitter data

Abstract

This work introduces a novel method for sentiment analysis in the stock market by combining Deep Neural Networks (DNN) and an improved Firefly Algorithm (FA). It is essential to comprehend investor sentiment in financial markets in order to forecast stock price movements and make wise investment choices. With the help of Deep Neural Networks, which are specially designed for sentiment analysis in stock market data, and the optimisation skills of the Firefly Algorithm, the suggested method seeks to increase classification accuracy. The effectiveness of the suggested approach is shown by means of empirical evaluations on benchmark datasets and a custom dataset created from actual stock market data. A comparative study with the most advanced techniques reveals significant improvements in sentiment emotion classification performance. The combination of FA and DNN presents a viable path forward for the development of sentiment analysis in the financial markets, enhancing our comprehension of stock market sentiments and enabling better-informed investment approaches.