E-SENTIMENTS AND STOCK MARKET PREDICTION: A BIBLIOMETRIC ANALYSIS

Purpose: The purpose of this study is to conduct a bibliometric analysis of research related to e-sentiments and stock market prediction. It aims to map the evolution of the field, identify key trends, and examine the methodologies used in sentiment-based stock market forecasting. The study also identifies influential authors, institutions, and countries contributing to the development of this research area.Design/Methodology/Approach: This research employs bibliometric analysis through tools such as the biblioshiny interface from the bibliometrix package in R-studio and VOS viewer software. The analysis includes data from publications between 2000 and 2025, focusing on citation networks, author collaborations, keyword co-occurrences, and the geographical distribution of research. The study provides insights into major trends, methodologies, and the growth of the e-sentiment analysis field in stock market prediction.Findings: The findings show significant growth in the number of publications, particularly from 2020 onwards, with a peak in 2024. Key contributors to this field include prominent journals like Expert Systems with Applications and IEEE Access, as well as leading authors such as Liang, C., and Ma, F. Research output is heavily concentrated in China, India, and the United States. The integration of machine learning techniques such as Long Short-Term Memory (LSTM) networks and Support Vector Machines (SVM) has been crucial in enhancing prediction accuracy. Despite the promising advancements, challenges like data quality, noise, and ambiguity in sentiment signals remain.Originality/Value: This study contributes to the understanding of e-sentiment analysis by providing a comprehensive overview of its evolution, identifying emerging trends, and highlighting the gaps in current research. It offers valuable insights into how sentiment analysis can enhance stock market prediction models by incorporating emotional and psychological factors that traditional models often overlook.