Open access
Aug 2026
Attention-enhanced LSTM framework for stock forecasting using Bayesian optimization
A hybrid framework by integrating Multi-Head Self-Attention mechanism into baseline LSTM architecture to dynamically weight critical historical features, combined with Bayesian optimization to systematically refine hyper-parameters for superior forecasting accuracy is proposed.
S. Shameem, Sonal S. Deshmukh
· Indian Journal of Science an... · 0 citations