Academic Corpus Analyzer

Deep Computational Linguistics Thesis Engine (Phase 1)

Comparative Corpus Evaluation: Human Baseline vs Pure AI vs AI-Humanized

Corpus A: Human Baseline

N/A

Experimental Vector Space Metric comparing relative stylistic signal proximity.

AWL Density

Syntactic Tree Depth

Readability Grade

Evaluates conceptual drift using Sentence-Transformers (MiniLM) embeddings.

Hum ↔ AI (A/B)

Hum ↔ Hum'd (A/C)

AI ↔ Hum'd (B/C)

Measures grammatical function-word Z-score divergence. Lower = stylistically closer.

Delta (A ↔ B)

Delta (A ↔ C)

Delta (B ↔ C)

Click metric name for technical info

Linguistic Diagnostic Feature	Human Baseline	Pure AI	AI-Humanized