Financial Market Analysis Using Deep Learning: A Framework for Market Behavior Prediction

The prediction of financial market trends has always been a challenging yet critical task due to the market's dynamic, nonlinear, and volatile nature. This paper proposes a deep learning-based framework for predicting stock price movements by leveraging historical time series data. The study evaluates the performance of different deep architectures including Long Short-Term Memory (LSTM), Bidirectional LSTM (BILSTM), and CNN-LSTM on financial datasets. Experimental results demonstrate that the hybrid CNN-LSTM model achieves the highest directional accuracy and lowest prediction error compared to standalone models. These findings highlight the effectiveness of deep learning in capturing temporal and structural patterns within financial data.

کلیدواژه ها:

Deep Learning ، LSTM ، Financial Markets ، Price Prediction ، CNN-LSTM ، Stock Forecasting