TY - GEN
T1 - Integrating Public Sentiment on Stock Return Percentage Prediction In Emerging Markets with A Deep Learning Approach
AU - Nugroho, Athaya Reyhan
AU - Sholikah, Rizka Wakhidatus
AU - Ciptaningtyas, Henning Titi
AU - Husni, Muchammad
AU - Indrawanti, Annisaa Sri
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - This study explores the influence of public sentiment on a particular stock that experienced a significant price increase despite having a high valuation. The research examines the impact of public sentiment on the percentage gain of this stock using a deep learning approach. Sentiment analysis is performed using BERT on data gathered from stock market applications, and the sentiment score are integrated to a return prediction model based on LSTM. The objective of this study is to determine the extent to which public sentiment influences stock price increases. Evaluation metrics used in this study includes R2, MSE, and RMSE. Results show that the LSTM model with sentiment input performs better (RMSE: 0.402, MSE: 0.634, R2: 0.523) than the model without (RMSE: 0.447, MSE: 0.669, R2: 0.470), indicating that public sentiment contributes positively to prediction accuracy.
AB - This study explores the influence of public sentiment on a particular stock that experienced a significant price increase despite having a high valuation. The research examines the impact of public sentiment on the percentage gain of this stock using a deep learning approach. Sentiment analysis is performed using BERT on data gathered from stock market applications, and the sentiment score are integrated to a return prediction model based on LSTM. The objective of this study is to determine the extent to which public sentiment influences stock price increases. Evaluation metrics used in this study includes R2, MSE, and RMSE. Results show that the LSTM model with sentiment input performs better (RMSE: 0.402, MSE: 0.634, R2: 0.523) than the model without (RMSE: 0.447, MSE: 0.669, R2: 0.470), indicating that public sentiment contributes positively to prediction accuracy.
KW - BERT (bidirectional encoder representative transformer)
KW - LSTM (long short-term memory)
KW - deep learning
KW - sentiment analysis
KW - time-series forecasting
UR - https://www.scopus.com/pages/publications/105012715799
U2 - 10.1109/SIML65326.2025.11081119
DO - 10.1109/SIML65326.2025.11081119
M3 - Conference contribution
AN - SCOPUS:105012715799
T3 - 2025 International Conference on Smart Computing, IoT and Machine Learning, SIML 2025
BT - 2025 International Conference on Smart Computing, IoT and Machine Learning, SIML 2025
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2025 International Conference on Smart Computing, IoT and Machine Learning, SIML 2025
Y2 - 3 June 2025 through 4 June 2025
ER -