A Comparative Analysis between Sentiment Analysis Methods in Stock Prediction of IoT-based companies

Abstract

Predictions in financial markets have attracted much of scholars' interest. Since the stock market is generally defined as dynamic and volatile due to the affect of the well-known "Internet of Things"(IoT). It is a fact that IoT significantly impacts the financial market and stocks in various ways. Therefore, numerous researchers have attempted to detect patterns regarding the way financial markets react to various external factors. In the last decade, the integration of machine learning and sentiment analysis has appeared to be an effective approach for improving IoT-based stock price predictions in the financial field. This paper intends to propose an approach involving financial news as a source of information for forecasting the stock movement of noteworthy IoT-related companies, employing Big Data Analytics (BDA), Machine Learning, and Sentiment Analysis methods as well. A comparative analysis is conducted between lexicon-based and machine learning-based methods of sentiment analysis about their performance, particularly in the domain of IoT-related stock market prediction.