Enhanced Stock Price Prediction Using Optimized Deep LSTM Model

G. Prathibha Priyadarshini, M. Sai Madhuri, T. Vishnu Priya, S. Moheeja, U. Lakshmi Prasanna

2025

Abstract

Stock price prediction is a challenging time- series task because the stock market is random and volatile. In this paper, we propose a better Deep Long Short-Term Memory (LSTM) network for accurate stock price prediction. The proposed model uses past stock attributes such as open, close, high, low, and volume, and technical indicators for predictive accuracy. For best performance, the hyperparameter optimization methods like Grid Search and Bayesian Optimization are used to fine-tune the best network structure. The model has multiple LSTM layers, dropout regularization to avoid overfitting, and adaptive learning rate optimizer to converge faster. Experiment results indicate that our enhanced Deep LSTM network performs superior to conventional machine learning methods and standard LSTM networks in Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). Our research enables better financial decision- making with accurate stock price forecasts for investors and traders.

Paper Citation

in Harvard Style

Priyadarshini G., Madhuri M., Priya T., Moheeja S. and Prasanna U. (2025). Enhanced Stock Price Prediction Using Optimized Deep LSTM Model. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 443-448. DOI: 10.5220/0013931100004919

in Bibtex Style

@conference{icrdicct`2525,
author={G. Priyadarshini and M. Madhuri and T. Priya and S. Moheeja and U. Prasanna},
title={Enhanced Stock Price Prediction Using Optimized Deep LSTM Model},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={443-448},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013931100004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Enhanced Stock Price Prediction Using Optimized Deep LSTM Model
SN - 978-989-758-777-1
AU - Priyadarshini G.
AU - Madhuri M.
AU - Priya T.
AU - Moheeja S.
AU - Prasanna U.
PY - 2025
SP - 443
EP - 448
DO - 10.5220/0013931100004919
PB - SciTePress