Imagen de Portada

High-Precision Stock Market Forecasting with Hybrid CNN-LSTM and Four Nature-Inspired Metaheuristics

Fuente: Zenodo
Guardado en:
Detalles Bibliográficos
Autores principales: Ruby Beniwal, Shruti Kalra
Formato: Recurso digital
Lenguaje:inglés
Publicado: Zenodo 2025
Materias:
Stock Market Forecasting, Hybrid CNN-LSTM, Metaheuristic Optimization, Deep Learning Models, Financial Time-Series Prediction
Acceso en línea:
Acceder al recurso
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866902206107615232
author Ruby Beniwal
Shruti Kalra
author_facet Ruby Beniwal
Shruti Kalra
contents <p><strong><span lang="EN-US">Abstract</span></strong></p> <p><span lang="EN-US">This<span> </span><span>paper</span><span> </span><span>presen</span><span>ts</span><span> </span>a<span> </span>high-precision<span> </span>stock<span> </span><span>market</span><span> </span>forecasting<span> </span><span>framework</span><span> </span><span>integrat-</span></span><span lang="EN-US"> </span><span lang="EN-US">ing<span> </span><span>CNN-LSTM</span><span> </span><span>h</span><span>ybrid</span><span> </span><span>net</span><span>works</span><span> </span>with<span> </span>four<span> </span>nature-inspired<span> </span>metaheuristic<span> </span>algo-</span><span lang="EN-US"> </span><span lang="EN-US">rithmsAnt</span><span lang="EN-US"> </span><span lang="EN-US">Colon</span><span lang="EN-US">y</span><span lang="EN-US"> </span><span lang="EN-US">Optimization<span> </span><span>(ACO),</span><span> </span><span>Particle</span><span> </span><span>Swarm</span><span> </span>Optimization<span> </span>(PSO),</span><span lang="EN-US"> </span><span lang="EN-US">Differen</span><span lang="EN-US">tial</span><span lang="EN-US"> </span><span lang="EN-US">Ev</span><span lang="EN-US">olution</span><span lang="EN-US"> </span><span lang="EN-US">(DE),<span> </span>and<span> </span>Artificial<span> </span>Bee<span> </span><span>Colon</span><span>y</span><span> </span>(ABC).<span> </span>Utilizing<span> </span>a<span> </span><span>14-year</span></span><span lang="EN-US"> </span><span lang="EN-US">dataset<span> </span>(20052019)<span> </span>from<span> </span><span>Y</span><span>ahoo</span><span> </span>Finance<span> </span><span>co</span><span>vering</span><span> </span>three<span> </span><span>major</span><span> </span><span>stoc</span><span>ks</span><span> </span>(Reliance,</span><span lang="EN-US"> </span><span lang="EN-US">TCS,<span> </span><span>HDF</span><span>C</span><span> </span>Bank),<span> </span>models<span> </span><span>achiev</span><span>ed</span><span> </span>an<span> </span>F1-Score<span> </span>of<span> </span>99.99%<span> </span>and<span> </span>R</span><strong><span lang="EN-US">2<span> </span></span></strong><span lang="EN-US">of<span> </span><span>99.98%</span><span>.</span></span><span lang="EN-US"> </span><span lang="EN-US">The<span> </span><span>h</span><span>ybrid</span><span> </span><span>model</span><span> </span><span>significantly</span><span> </span>outperformed<span> </span>standalone<span> </span>ANN,<span> </span>RNN,<span> </span>LSTM,<span> </span>and</span><span lang="EN-US"> </span><span lang="EN-US">CNN<span> </span><span>architectures.</span><span> </span>Experimental<span> </span>results<span> </span>confirm<span> </span>the<span> </span>robustness<span> </span>of<span> </span>deep<span> </span>learning</span><span lang="EN-US"> </span><span lang="EN-US">in<span> </span>financial<span> </span>forecasting,<span> </span>demonstrating<span> </span>superior<span> </span>accuracy<span> </span>(=99%)<span> </span>and<span> </span>enhanced</span><span lang="EN-US"> </span><span lang="EN-US">feature <span>sele</span><span>ction</span><span> </span>efficiency <span>(=95</span><span>%</span><span>)</span><span> </span>through optimization <span>techniques.</span></span></p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_17278357
institution Zenodo
language eng
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle High-Precision Stock Market Forecasting with Hybrid CNN-LSTM and Four Nature-Inspired Metaheuristics
Ruby Beniwal
Shruti Kalra
Stock Market Forecasting, Hybrid CNN-LSTM, Metaheuristic Optimization, Deep Learning Models, Financial Time-Series Prediction
<p><strong><span lang="EN-US">Abstract</span></strong></p> <p><span lang="EN-US">This<span> </span><span>paper</span><span> </span><span>presen</span><span>ts</span><span> </span>a<span> </span>high-precision<span> </span>stock<span> </span><span>market</span><span> </span>forecasting<span> </span><span>framework</span><span> </span><span>integrat-</span></span><span lang="EN-US"> </span><span lang="EN-US">ing<span> </span><span>CNN-LSTM</span><span> </span><span>h</span><span>ybrid</span><span> </span><span>net</span><span>works</span><span> </span>with<span> </span>four<span> </span>nature-inspired<span> </span>metaheuristic<span> </span>algo-</span><span lang="EN-US"> </span><span lang="EN-US">rithmsAnt</span><span lang="EN-US"> </span><span lang="EN-US">Colon</span><span lang="EN-US">y</span><span lang="EN-US"> </span><span lang="EN-US">Optimization<span> </span><span>(ACO),</span><span> </span><span>Particle</span><span> </span><span>Swarm</span><span> </span>Optimization<span> </span>(PSO),</span><span lang="EN-US"> </span><span lang="EN-US">Differen</span><span lang="EN-US">tial</span><span lang="EN-US"> </span><span lang="EN-US">Ev</span><span lang="EN-US">olution</span><span lang="EN-US"> </span><span lang="EN-US">(DE),<span> </span>and<span> </span>Artificial<span> </span>Bee<span> </span><span>Colon</span><span>y</span><span> </span>(ABC).<span> </span>Utilizing<span> </span>a<span> </span><span>14-year</span></span><span lang="EN-US"> </span><span lang="EN-US">dataset<span> </span>(20052019)<span> </span>from<span> </span><span>Y</span><span>ahoo</span><span> </span>Finance<span> </span><span>co</span><span>vering</span><span> </span>three<span> </span><span>major</span><span> </span><span>stoc</span><span>ks</span><span> </span>(Reliance,</span><span lang="EN-US"> </span><span lang="EN-US">TCS,<span> </span><span>HDF</span><span>C</span><span> </span>Bank),<span> </span>models<span> </span><span>achiev</span><span>ed</span><span> </span>an<span> </span>F1-Score<span> </span>of<span> </span>99.99%<span> </span>and<span> </span>R</span><strong><span lang="EN-US">2<span> </span></span></strong><span lang="EN-US">of<span> </span><span>99.98%</span><span>.</span></span><span lang="EN-US"> </span><span lang="EN-US">The<span> </span><span>h</span><span>ybrid</span><span> </span><span>model</span><span> </span><span>significantly</span><span> </span>outperformed<span> </span>standalone<span> </span>ANN,<span> </span>RNN,<span> </span>LSTM,<span> </span>and</span><span lang="EN-US"> </span><span lang="EN-US">CNN<span> </span><span>architectures.</span><span> </span>Experimental<span> </span>results<span> </span>confirm<span> </span>the<span> </span>robustness<span> </span>of<span> </span>deep<span> </span>learning</span><span lang="EN-US"> </span><span lang="EN-US">in<span> </span>financial<span> </span>forecasting,<span> </span>demonstrating<span> </span>superior<span> </span>accuracy<span> </span>(=99%)<span> </span>and<span> </span>enhanced</span><span lang="EN-US"> </span><span lang="EN-US">feature <span>sele</span><span>ction</span><span> </span>efficiency <span>(=95</span><span>%</span><span>)</span><span> </span>through optimization <span>techniques.</span></span></p>
title High-Precision Stock Market Forecasting with Hybrid CNN-LSTM and Four Nature-Inspired Metaheuristics
topic Stock Market Forecasting, Hybrid CNN-LSTM, Metaheuristic Optimization, Deep Learning Models, Financial Time-Series Prediction
url https://doi.org/10.5281/zenodo.17278357

Ejemplares similares