| name | evez-eigenforensics |
| description | Spectral gap detection in document corpora. Turns FOIA documents into reference graphs and finds what they're hiding via negative eigenvalues. Found 5 gaps in the AARO UAP report at p<0.05. Use for FOIA analysis, censorship detection, document gap analysis. |
| version | 1.0.0 |
| author | @EvezArt |
| tags | ["evez","eigenforensics","foia","uap","aaro","spectral","censorship","gap-detection"] |
EVEZ Eigenforensics Engine
Documents are graphs. Structural holes are negative eigenvalues. What they omit is louder than what they say.
What It Does
- Parses document corpora into reference graphs
- Computes eigenvalue spectra of the document adjacency matrix
- Detects structural holes (negative eigenvalues) = what's being hidden
- Statistical significance via permutation testing (p<0.05)
Key Result
Applied to AARO (All-domain Anomaly Resolution Office) UAP report:
- 5 gaps found at p<0.05
- Dominant negative eigenvalue λ = -0.333 (37% of system tension)
- Censorship IS the dominant negative eigenvalue
Falsifiable
If removing >5% of documents from a corpus does NOT produce eigenvalue shifts detectable at p<0.05, the theorem is false.
Quick Start
from spectral import EigenforensicsEngine
engine = EigenforensicsEngine()
engine.load_corpus("./foia_documents/")
gaps = engine.detect_gaps()
print(f"{len(gaps)} gaps found at p<0.05")
Author
Steven Crawford-Maggard (EVEZ666)
License: MIT
