Saved in:
Bibliographic Details
Main Authors: Madhusmita Behera, Tripurari kumar, Biswajit Bej, Yanamadni Venkata Sasank Kumar
Format: Recurso digital
Language:
Published: Zenodo 2025
Online Access:https://doi.org/10.5281/zenodo.15401754
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866901965419577344
author Madhusmita Behera
Tripurari kumar
Biswajit Bej
Yanamadni Venkata Sasank Kumar
author_facet Madhusmita Behera
Tripurari kumar
Biswajit Bej
Yanamadni Venkata Sasank Kumar
contents <p> Abstract—Stock price prediction is a complex challenge in<br>f<br> luenced by various factors, including historical trends and<br> market sentiment. Existing models often rely on single-source<br> data, limiting their forecasting accuracy. This study presents<br> a hybrid approach that integrates Long Short-Term Memory<br> (LSTM) networks for time-series forecasting with sentiment<br> analysis derived from financial news. To dynamically combine<br> the two prediction outputs, Bayesian Optimization is employed<br> to calculate optimal weights, ensuring adaptability to market<br> f<br> luctuations. Experimental results demonstrate that the proposed<br> method achieves improved accuracy, reducing the Mean Squared<br> Error compared to traditional fixed-weight models. This research<br> contributes to the field of financial forecasting by providing a<br> more responsive and data-driven prediction framework</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_15401754
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Stock Price Prediction Using LSTM and Sentiment Analysis
Madhusmita Behera
Tripurari kumar
Biswajit Bej
Yanamadni Venkata Sasank Kumar
<p> Abstract—Stock price prediction is a complex challenge in<br>f<br> luenced by various factors, including historical trends and<br> market sentiment. Existing models often rely on single-source<br> data, limiting their forecasting accuracy. This study presents<br> a hybrid approach that integrates Long Short-Term Memory<br> (LSTM) networks for time-series forecasting with sentiment<br> analysis derived from financial news. To dynamically combine<br> the two prediction outputs, Bayesian Optimization is employed<br> to calculate optimal weights, ensuring adaptability to market<br> f<br> luctuations. Experimental results demonstrate that the proposed<br> method achieves improved accuracy, reducing the Mean Squared<br> Error compared to traditional fixed-weight models. This research<br> contributes to the field of financial forecasting by providing a<br> more responsive and data-driven prediction framework</p>
title Stock Price Prediction Using LSTM and Sentiment Analysis
url https://doi.org/10.5281/zenodo.15401754