This document discusses using artificial neural networks (ANN) to predict stock market prices. It proposes a new ANN model for feature extraction and selection to more accurately predict stock exchange markets. The model involves preprocessing noisy stock data using technical indicators, extracting and selecting key features, then applying ANN and data mining techniques to accurately predict stock prices. Evaluation of the ANN model found that feature extraction and selection play an important role in the robustness and efficiency of predicting stock market movement.