Machine Learning Models in Finance

We often say "AI", but that's a buzzword. Specifically, TradingMaster uses a hybrid ensemble of Machine Learning (ML) models.

1. LSTM (Long Short-Term Memory)

• What it does: It remembers sequences.
• Use Case: Recognizing chart patterns. It knows that Pattern A usually leads to Result B because it has seen it 50,000 times before.

2. Random Forest

• What it does: It creates thousands of "Decision Trees" (If X, then Y) and averages them.
• Use Case: Classification. "Is this market Bullish or Bearish?" It prevents overfitting to one specific indicator.

3. NLP (Natural Language Processing)

• What it does: Reads text and understands emotion.
• Use Case: Sentiment Analysis. Scanning headlines for keywords that historically crash the market.

Why Hybrid?

No single model is perfect. By voting across multiple models (Ensemble Learning), we reduce the error rate significantly. If the LSTM says "Buy" but the Random Forest says "Sell", the Confidence Score drops to 50% (neutral), keeping you safe.