Correlation Network
Visualize relationships between insider trades, related news articles, and SEC filings to identify potential patterns of suspicious activity.
Why This Graph Matters
Traditional insider trading detection looks at trades in isolation. Our network graph reveals hidden connections — when an executive sells before bad news, or multiple insiders trade around the same corporate event. The denser the connections around a trade, the more likely it warrants investigation. This visual approach has helped analysts identify patterns that spreadsheet analysis misses.
How to Read This Graph
Node Types
Data Sources
Connections
Line thickness = correlation strength
Thicker lines = stronger relationship
Lines connect trades to related news and filings
Red Flags 🚩
• Clusters of red trade nodes
• Trades with many news connections
• Dense filing correlations
• Isolated high-risk trades
Understanding the AI Analysis
News Correlation
Our AI uses FinBERT sentiment analysis to compare insider trade timing with related news. High correlation (thick cyan lines) suggests the insider may have acted on material non-public information.
Filing Correlation
SEC Form 4 filings are analyzed using semantic similarity to detect patterns. Strong correlations (thick orange lines) may indicate coordinated insider activity or unusual filing patterns.