AlphaLeap

AI Explainability Dashboard

Understanding how AlphaLeap's AI makes investment recommendations.

Data Sources
Data Sources

Our AI processes a vast array of data, including:

  • Traditional financial statements (P/E, P/B, FCF)
  • Market data (price, volume, volatility)
  • Alternative data (news sentiment, social media trends, job postings, patent filings, web traffic, etc.)
  • Macroeconomic indicators
Core Methodologies
Methodologies

The AI integrates established and innovative strategies:

  • Value Investing: Identifying undervalued companies (e.g., Greenblatt's Magic Formula).
  • Growth Investing: Targeting companies with high growth potential.
  • Factor Investing: Focusing on drivers of return (Value, Size, Momentum, Quality).
  • Behavioral Finance: Exploiting market anomalies from psychological biases.
Machine Learning Models
Machine Learning

Sophisticated algorithms power our predictions:

  • Ensemble methods (Random Forest, XGBoost)
  • Deep Learning (LSTMs for time series)
  • Natural Language Processing (NLP) for sentiment

Models are continuously trained and validated.

Key Factor Influence (Illustrative)
For each stock recommendation, the AI weighs various factors. The "Rationale" provided with each AI-driven output details the specific reasoning. Below is a general idea of factors considered:

Example Stock: TECHCORP (TCORP) - AI Recommendation: BUY

Rationale Snippet: "TECHCORP shows strong revenue growth (Factor: Growth Investing), positive social media sentiment (Factor: Alternative Data), and a favorable P/E ratio compared to peers (Factor: Value Investing). Recent patent filings indicate ongoing innovation (Factor: Alternative Data - Patent). Job postings have increased by 12% (Factor: Alternative Data - Job Postings), signaling expansion."

The AI's rationale for each specific stock pick (available in the AI Stock Selection, Alternative Data, and Daily Filings Summary sections) explains the primary drivers for that particular recommendation. This includes how financial metrics, alternative data signals, and model interpretations contribute to the final decision.

We aim for transparency by providing these rationales, helping you understand the "why" behind AI's suggestions.

Our Commitment to Transparency

While AI models can be complex, we strive to make their decisions understandable. The "Rationale" accompanying each AI-generated output highlights the key factors influencing the recommendation. Our goal is to empower you with insights, not just black-box predictions. Continuous research is underway to enhance explainability further using techniques like SHAP analysis.