MLKN.lab Hierarchy Master File and Knowledge Network Data (All Layers, All Details)

MLKN.lab Hierarchy Master File and Knowledge Network Data (All Layers, All Details) Authors/Creators : Papin, François (Researcher)URL : [https://francoispapin.github.io/MLKN-lab/](https://francoispapin.github.io/MLKN-lab/) --- How to Cite If you use this dataset, please cite it as: APA Style: Papin, F. (2026). MLKN.lab Hierarchy Master File and Knowledge Network Data (All Layers, All Details) [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.21363227 BibTeX: ```bibtex@dataset{papin2026mlkn, author = {Papin, François}, title = {MLKN.lab Hierarchy Master File and Knowledge Network Data (All Layers, All Details)}, year = {2026}, publisher = {Zenodo}, doi = {10.5281/zenodo.21363227}, url = {https://doi.org/10.5281/zenodo.21363227}} --- Description A comprehensive dataset for MLKN.lab's polyhierarchical knowledge network, containing all layers and details of the hierarchy, as well as JSON files for interactive visualization. MLKN.lab (Multi-Layered Knowledge Network Ideas Laboratory) is a research project that models scientific knowledge as a polyhierarchical hypergraph. The dataset includes the complete, multi-layered hierarchy of scientific domains, subdomains, topics, and concepts, extracted from [OpenAlex](https://docs.openalex.org/) and structured for network analysis, visualization, and computational epistemology. --- Dataset Overview This deposit includes four core files for the MLKN.lab project: 1. [MLKN_Hierarchy_Master_File_All_layers_All_Details_Final.csv](MLKN_Hierarchy_Master_File_All_layers_All_Details_Final.csv)- Format: CSV (Comma-Separated Values)- Size: 92.9 MB- Rows: 326,862- Columns: 10 (after removing empty columns)- Layers: 5 ontological layers (Core Domains → Academic Disciplines → Academic Subdisciplines → Core Thematic Domains → Main Concepts)- Source: Derived from [OpenAlex API](https://docs.openalex.org/) (free, open catalog of scholarly papers).- Last Updated: June 2026- Purpose: Master hierarchy file for network analysis, meta-science research, and computational epistemology. ---CSV Schema (`MLKN_Hierarchy_Master_File_All_layers_All_Details_Final.csv`) | Column | Type | Description | Example | |----------------------------------|--------- |-------------------------------------------------------------------------------|-------------------------------------| | `MLKN_Core_Domain` | String | High-level domain (e.g., "Social Sciences & Humanities"). | "Social Sciences & Humanities" | | `OpenAlex_Core_Domain` | String | Corresponding OpenAlex domain. | "Social Sciences & Humanities" | | `Academic_Discipline_Name` | String | Name of the academic discipline (e.g., "Psychology"). | "Psychology" | | `Academic_Subdiscipline_Name` | String | Name of the subdiscipline (e.g., "General Psychology"). | "General Psychology" | | `Core_Thematic_Domain_Name` | String | Thematic domain name (e.g., "Academic and Historical Perspectives in Psychology"). | "Academic and Historical Perspectives" | | `Core_Thematic_Domain_ID` | String | OpenAlex ID for the thematic domain. | "https://openalex.org/T12430" | | `Main_Concept_Name` | String | Name of the main concept (e.g., "Computer science"). | "Computer science" | | `Main_Concept_ID` | String | OpenAlex ID for the concept. | "https://openalex.org/T12430" | | `Source_File` | String | Source file used for classification. | "MLKN_Hierarchy_Reclassified_Core_Domains.csv" | | `Notes` | String | Additional notes (e.g., "Reclassified hierarchy with 6 Core Domains"). | "Reclassified hierarchy with 6 Core Domains" | ---2. [MLKN_hypergraph_nodes.json](MLKN_hypergraph_nodes.json)- Format: JSON (JavaScript Object Notation)- Size: 5.6 MB- Nodes: 31,590- Structure: Each node includes `id`, `name`, `Layer`, `Core Domain`, and metadata for visualization.- Purpose: Node data for the [MLKN.hypergraph](https://francoispapin.github.io/MLKN-lab/knowledge_network/MLKN-hypergraph.html) interactive visualization (D3.js). ---3. [MLKN_hypergraph_edges.json](MLKN_hypergraph_edges.json)- Format: JSON (JavaScript Object Notation)- Size: 40.5 MB- Edges: 320,075- Structure: Each edge includes `source`, `target`, `type`, and `weight` for network connections.- Purpose: Edge data for the [MLKN.hypergraph](https://francoispapin.github.io/MLKN-lab/knowledge_network/MLKN-hypergraph.html) interactive visualization (D3.js). ---4. [MLKN_tutorial.ipynb](MLKN_tutorial.ipynb)- Format: Jupyter Notebook- Purpose: Quickstart guide with examples for loading the CSV, converting it to a NetworkX graph, and calculating basic metrics (e.g., degree centrality). --- Ontology Standards and Methodological Foundations The MLKN.lab hierarchy is built upon a reclassified ontology that extends and refines the original OpenAlex taxonomy (4 Core Domains) to 6 Core Domains, aligning with both SKOS (Simple Knowledge Organization System) principles and the OECD Frascati Classification standards. This reclassification addresses a critical gap in OpenAlex's original structure, where Physical Sciences grouped together disciplines with distinct epistemological foundations (e.g., mathematics, computer science, physics, and engineering). 🔹 Reclassification DetailsOpenAlex's original 4 Core Domains:1. Physical Sciences (Physics, Mathematics, Computer Science, Engineering)2. Life Sciences (Biology, Agriculture)3. Health Sciences (Medicine, Public Health)4. Social Sciences & Humanities (Psychology, Sociology, Arts) MLKN.lab's 6 Core Domains (aligned with Frascati and SKOS):1. Formal Sciences (Mathematics, Theoretical Computer Science, Logic) - Rationale: Formal Sciences are foundational to all other domains (e.g., mathematics underpins physics, computer science, and statistics). Their explicit separation enables deeper studies of knowledge diffusion (e.g., how formal methods influence applied sciences).2. Natural Sciences (Physics, Chemistry, Earth Sciences) - Rationale: Focuses on empirical study of the natural world, distinct from formal abstraction or engineering applications.3. Engineering and Technology (Computer Science, Engineering, Applied Mathematics) - Rationale: Highlights applied and technological disciplines, often bridging Formal and Natural Sciences.4. Life Sciences (Biology, Medicine, Agriculture) - Rationale: Retained from OpenAlex, as it represents a coherent epistemological domain.5. Health & Medical Sciences (Medicine, Public Health, Clinical Research) - Rationale: Separated from Life Sciences to reflect its unique societal impact and methodological approaches.6. Social Sciences & Humanities (Psychology, Sociology, Arts, Philosophy) - Rationale: Preserves OpenAlex's original grouping, while enabling interdisciplinary studies (e.g., cognitive psychology's links to neuroscience or AI). 🔹 Theoretical and Methodological Implications- SKOS Alignment: Our hierarchy follows SKOS principles (e.g., hierarchical relationships via `skos:broader`/`skos:narrower`), enabling semantic interoperability with other knowledge organization systems (e.g., Wikidata, DBpedia).- Frascati Alignment: The 6-domain structure aligns with the OECD Frascati Manual, a global standard for classifying scientific disciplines, ensuring compatibility with international research policies.- Balanced Representation: By separating Formal Sciences, we address OpenAlex's underrepresentation of foundational disciplines, allowing for more accurate analyses of knowledge flows (e.g., how mathematical theories diffuse into physics or engineering).- Interdisciplinarity: This reclassification reveals hidden bridges between domains (e.g., Formal Sciences ↔ Engineering, or Social Sciences ↔ Health Sciences), which are critical for studying scientific convergence and innovation pathways. ---*This ontological refinement is not merely technical—it reflects a philosophical commitment to representing scientific knowledge in a way that is both rigorous and *actionable for meta-science research. --- Key Statistics (Data Sharing Standards) | Metric | Master File (CSV) | Nodes (JSON) | Edges (JSON) ||------------------------------|---------------------- |------------------- |------------------|| Format | CSV | JSON | JSON || Size | 92.9 MB | 5.6 MB | 40.5 MB || Records/Elements | 326,862 rows | 31,590 nodes | 320,075 edges || Layers | 5 | 5 | 5 || Core Domains | 6 | 6 | N/A || Academic Disciplines | 25 | 25 | N/A || Academic Subdisciplines | 235 | 235 | N/A || Core Thematic Domains | 4,229 | 4,229 | N/A | | Main Concepts | 27,095 | 27,095 | N/A || Source | OpenAlex | OpenAlex | OpenAlex || Last Updated | June 2026 | June 2026 | June 2026 || Optimized For | Analysis | Visualization | Visualization | *Counts for Core Thematic Domains and Main Concepts are approximate due to filtering for visualization optimization.* --- Intended Use Cases For the Master File (CSV):1. Network Analysis: - Study the topological structure of scientific knowledge (e.g., using [NetworkX](https://networkx.org/), [igraph](https://igraph.org/), or [Gephi](https://gephi.org/)). - Identify key disciplines, bridges between fields, or emerging research areas. 2. Meta-Science Research: - Analyze interdisciplinarity, knowledge diffusion, or the evolution of scientific fields. - Study **how disciplines connect, overlap, or diverge** over time. 3. Computational Epistemology: - Model how knowledge is structured, validated, and evolved in computational systems. - Develop AI systems that reason over scientific knowledge. ---For the JSON Files (Nodes & Edges):1. Interactive Visualization: - Power the [MLKN.hypergraph](https://francoispapin.github.io/MLKN-lab/knowledge_network/MLKN-hypergraph.html) visualization (D3.js). - Explore polyhierarchical relationships between domains in a user-friendly interface. 2. Web Applications: - Integrate into knowledge management tools or educational platforms. - Build custom network explorers for specific research questions. 3. Prototyping: - Test new algorithms for network analysis or knowledge mapping. - Develop demos for presentations or papers. --- Example Use Case: Identifying Interdisciplinary Bridges Using the CSV file, you can identify bridges between disciplines by analyzing nodes that appear in multiple `Core_Thematic_Domain_Name` categories. For example, the concept "Machine Learning" appears in both "Computer Science" and "Mathematics", suggesting a strong interdisciplinary link. Python Example: ```pythonimport pandas as pdimport networkx as nx # Load the CSVdf = pd.read_csv("MLKN_Hierarchy_Master_File_All_layers_All_Details_Final.csv") # Find concepts that appear in multiple thematic domainsduplicate_concepts = df['Main_Concept_Name'].value_counts()interdisciplinary_concepts = duplicate_concepts[duplicate_concepts > 1].index.tolist()print("Interdisciplinary concepts:", interdisciplinary_concepts[:10]) # Top 10 --- Data Source OpenAlex Snapshot Date: March 2026 (raw data snapshot from OpenAlex). Data Processing Date: June, 15 (hierarchy construction and cleaning). Ontology Standards: The MLKN.lab hierarchy follows SKOS (Simple Knowledge Organization System) principles, with a reclassified ontology of 6 Core Domains (instead of OpenAlex's 4) to achieve a more balanced representation of scientific knowledge. OpenAlex Version: [OpenAlex Documentation](https://docs.openalex.org/). --- Data Consistency Notes The CSV master file and JSON files are derived from the same OpenAlex source but optimized for different use cases:- CSV: Complete hierarchy for analysis (includes all columns, with empty columns removed for clarity).- JSON: Lightweight files for web-based visualization (filtered columns, optimized structure).- Minor differences in row counts (e.g., duplicate nodes in CSV for hierarchical structure) are intentional to ensure optimal performance for the visualization.- All files are aligned with the MLKN.lab v1.0 hierarchy and compatible with each other. --- Related Resources - MLKN.lab Website: [https://francoispapin.github.io/MLKN-lab/](https://francoispapin.github.io/MLKN-lab/) Explore the project’s mission, methodology, and applications.- MLKN.hypergraph Visualization: [https://francoispapin.github.io/MLKN-lab/knowledge_network/MLKN-hypergraph.html](https://francoispapin.github.io/MLKN-lab/knowledge_network/MLKN-hypergraph.html) Interactive exploration of the knowledge network.- Scientific References: [https://francoispapin.github.io/MLKN-lab/references/references.html](https://francoispapin.github.io/MLKN-lab/references/references.html) A curated bibliography of 165+ references that inspire MLKN.lab.- GitHub Repository: [https://github.com/FrancoisPapin/MLKN-lab](https://github.com/FrancoisPapin/MLKN-lab) Access the code, documentation, and additional datasets for MLKN.lab. --- Citation If you use this dataset, please cite it as: Papin, F. (2026). MLKN.lab Hierarchy Master File and Knowledge Network Data (All Layers, All Details) [Dataset]. Zenodo. [https://doi.org/10.5281/zenodo.21363227](https://doi.org/10.5281/zenodo.21363227) --- Attached Files 1. [MLKN_Hierarchy_Master_File_All_layers_All_Details_Final.csv](MLKN_Hierarchy_Master_File_All_layers_All_Details_Final.csv) (Master hierarchy)2. [MLKN_hypergraph_nodes.json](MLKN_hypergraph_nodes.json) (Nodes for MLKN.hypergraph)3. [MLKN_hypergraph_edges.json](MLKN_hypergraph_edges.json) (Edges for MLKN.hypergraph)4. [MLKN_tutorial.ipynb](MLKN_tutorial.ipynb) (Jupyter Notebook tutorial)5. [README.md](README.md) (Detailed usage instructions) --- Changelog - v1.0 (July 2026): Based on OpenAlex snapshot from March 2026, processed and validated on June 15, 2026. Removed empty columns, added schema documentation, and included a Jupyter tutorial. --- Update Cycle This dataset is a static snapshot of the OpenAlex database as of June 15, 2026. Future updates will be released as new versions on Zenodo, with a planned frequency of once per year (or upon major OpenAlex updates). ---License This dataset is licensed under the MIT License. You are free to use, copy, modify, and distribute the data for any purpose, including commercial use, as long as you include the citation above.