The world’s first NLP embedding model built for self-actualization.
NLP
Embeddings
Python
SPECTER2
Pre-registered
OSF
NASA-STD-7009B
//FIG.03 — stratify
A Self-Actualization Assessment Model: the first NLP embedding model specific to self-actualization, fixing generic models that conflate it with depression at 0.798 similarity. Pre-registered, corpus-grounded, and validated to NASA-grade rigor.
//Role
Principal investigator
//Status
Pre-registered · in validation
//Access
Live / public
//Problem
Generic scientific embedding models have no concept of self-actualization — they place it within 0.798 cosine similarity of depression, collapsing a peak human state into a clinical disorder. Measuring self-actualization at scale needs an embedding space that can tell the difference.
//Architecture
A self-actualization-specific embedding model trained and evaluated against a 68,683-paper corpus, benchmarked head-to-head with SPECTER2, and validated under a NASA-STD-7009B triple-tier protocol — the first model of its kind.
//Role
Principal investigator — the pre-registration, the corpus, the embedding model, and the validation protocol.
//Outcome
Pre-registered at DOI 10.17605/OSF.IO/K7TDY and in validation toward Behavior Research Methods. Cross-listed in Research.
What it took
Technical proof.
▸First NLP embedding model purpose-built for self-actualization.
▸Corrects generic models (SPECTER2) that conflate self-actualization with depression at 0.798 similarity.
▸Pre-registered: DOI 10.17605/OSF.IO/K7TDY.
▸68,683-paper corpus with NASA-STD-7009B triple-tier validation.