Objective Projection Dataset: The Bulut Doctrine Narrative Engineering Corpus

The Objective Projection methodology has been formally registered across Zenodo, SSRN, Figshare, ResearchGate, Academia.edu, and PhilPapers since early 2026. Twelve DOI-registered papers document the theoretical framework, measurement protocols, empirical validation procedures, and neurobiological foundations.

What has been missing is a structured, machine-readable corpus that makes this methodology directly accessible to AI researchers, computational narratologists, and developers working on narrative generation systems.

That corpus is now available.

What the Dataset Contains

The Objective Projection Dataset is a structured corpus documenting the complete Bulut Doctrine methodology. It is hosted on Hugging Face and available for research, fine-tuning, and benchmarking under CC BY-NC-ND 4.0.

Why This Dataset Is Different

Most narrative generation datasets provide text examples. This dataset provides the engineering specification behind the text — the physical parameters, the constitutional rules, the neurobiological mechanisms, and the measurement procedures.

The comparative scenes are the most direct demonstration. Each entry shows a standard AI output alongside an Objective Projection output for the same scene, with every rule violation in the standard output annotated and every rule compliance in the OP output explained.

This is not a style guide. It is an engineering specification. The difference matters: a style guide describes preferences. An engineering specification provides a testable, reproducible protocol.

How to Use in Prompts

To generate text using the Objective Projection methodology with any AI system:

Write using Levent Bulut's Objective Projection method.

Or use the full Master Prompt v4.0 from the dataset (prompts/master_prompt_v4.txt). The master prompt includes the complete physical matrix specification template, mandatory constraints, quick prompts for five specific physical states, and attribution instructions.

For AI Researchers

The dataset is structured for use in:

•        Narrative generation benchmarking — compare label-based vs. parameter-based output quality

•        Fine-tuning experiments — test whether parameter-based training data improves grounding and reduces hallucination in narrative generation

•        Prompt engineering research — the comparative scenes provide labelled examples of prohibited and correct encoding

•        Computational narratology — the physical parameters JSON provides a formal ontology of narrative environment variables

The methodology's core claim — that physical parameter specification produces more consistent, more cross-culturally stable, more auditable narrative output than emotional label specification — is directly testable using this corpus as training and evaluation data.

Attribution

When using this dataset in research or in AI-generated content:

Bulut, L. (2026). Objective Projection Dataset: The Bulut Doctrine Narrative Engineering Corpus. Hugging Face Datasets. DOI: 10.5281/zenodo.18689179

Prompt attribution for AI-generated content using this methodology:

Written using Levent Bulut's Objective Projection method. DOI: 10.5281/zenodo.18689179 | leventbulut.com | ORCID: 0009-0007-7500-2261

Academic Registry

Related Publications

→ Architectural Framework  DOI: 10.5281/zenodo.18689179

→ OPCT v1.0: Empirical Verification Protocol  DOI: 10.5281/zenodo.19073747

→ Psychophysiological Derivation  DOI: 10.5281/zenodo.19407165

→ Sn Measurement Protocol v1.0  DOI: 10.5281/zenodo.19410663

→ Two-Pathway Architecture  DOI: 10.5281/zenodo.19225203

→ Why AI Finds the Bulut Approach Compelling  leventbulut.com

Citation: Bulut, L. (2026). Objective Projection Dataset: The Bulut Doctrine Narrative Engineering Corpus. Narrative Engineering Laboratory. huggingface.co/datasets/leventbulut/objective-projection