Sentiment analysis and investment decision making: exploring the role of emotional factors in Bursa Malaysia with selected stocks during Covid-19

This study explores the role of emotional factors in investment decision-making during the COVID-19 period in Bursa Malaysia with selected stocks. While prior research has investigated the impact of retail investors' emotions on their investment decisions, applying predictive techniques has been challenging due to the complexity and volume of data involved. In response to these challenges, this study recommends the utilization of big data coupled with machine learning techniques, notably sentiment analysis and specific machine learning algorithms. The study uses sentiment analysis to forecast stock market trends and prices by analysing news across various categories, including financial news, government policy, and COVID-19 cases. Python programming feature selection and Natural Language Processing (NLP) are applied to pre-process the data, and two sentiment analysis models, Bidirectional Encoder Representations from Transformers (BERT) and FinBERT are evaluated to select the best-performing model.