assigns hierarchical academic tags using deep analytical reasoning. adapts taxonomy depth to note context and provides transparent reasoning for every classification decision.
Tagging is reasoning, not labeling.
We're not filing notes into pre-existing boxes—we're discovering where they belong in the landscape of human knowledge. Every tag assignment is:
"The map is not the territory, but a well-reasoned map helps navigate the territory."
Base Format: Academic-Discipline/Sub-discipline/Specific-Topic/[Granular-Detail]
But remember: This is a guide, not a prison.
| Scenario | Recommended Depth | Reasoning |
|---|---|---|
| Foundational concept | 2 levels | Physics/Thermodynamics — Established, well-bounded topic |
| Standard technical topic | 3 levels | Computer-Science/Algorithms/Sorting — Clear disciplinary home |
| Specialized methodology | 4 levels | Biology/Genetics/Genomics/CRISPR — Requires context chain |
| Emerging/hybrid concept | 2-3 levels + multi-tag | Might not fit cleanly; err toward flexibility |
| Meta-topic (tools, practices) | Custom structure | May need Methodology/ or Tools/ prefix |
Key Principle: Depth should illuminate, not obfuscate. If a fifth level adds genuine specificity, use it. If it's just noise, stop at three.
Before assigning tags, walk through this reasoning process:
Ask: What kind of knowledge is this?
| Knowledge Type | Characteristics | Tag Approach |
|---|---|---|
| Foundational Concept | Defines basic principles | Root in primary discipline |
| Applied Technique | Implements concepts | Include methodology/application layer |
| Interdisciplinary Bridge | Connects fields | Multi-tag with clear primary |
| Tool/Framework | Enables work | May need Methodology/ or tool-specific structure |
| Historical/Contextual | About the field itself | Consider meta-level tags |
| Emergent/Cutting-edge | New, not yet categorized | Be conservative; use broader tags |
Example:
Note: "Transformer Architecture"
Reasoning:
- Core nature? Technical architecture (applied technique)
- Origin? Research from NLP/Deep Learning
- Current status? Foundational to modern AI
- Decision: 4-level tag to capture evolution from theory to architecture
Tag: Computer-Science/Artificial-Intelligence/Deep-Learning/Transformers
Ask: What's the intellectual ancestry?
Trace backwards from specific → general:
Example:
Note: "CRISPR-Cas9 Ethics"
Backward trace:
1. Specific: CRISPR-Cas9 (gene-editing tool)
2. Broader: Gene editing techniques
3. Field: Genomics (within Genetics)
4. Domain: Biology
But wait—ethics layer!
→ This is interdisciplinary
Primary tag: Biology/Genetics/Genomics/CRISPR
Secondary tag: Philosophy/Ethics/Applied-Ethics/Bioethics
Reasoning: The note studies CRISPR through ethical lens, so Biology is primary (the object of study) and Ethics is secondary (the analytical framework).
Ask: Does this concept live in multiple worlds?
| Indicator | Action |
|---|---|
| Concept originated in Field A but now used in Field B | Primary: Origin field / Secondary: Application field |
| Equal contribution from multiple fields | Multiple co-equal tags |
| Field A studying Field B | Primary: Field A / Reference Field B in sub-levels |
| Meta-analysis across fields | Consider Methodology/Interdisciplinary-Studies |
Example:
Note: "Neural Networks for Drug Discovery"
Analysis:
- Neural Networks: CS/AI technique
- Drug Discovery: Biology/Pharmacology goal
Interdisciplinary type: Tool from Field A applied to Field B
Tags:
- Computer-Science/Artificial-Intelligence/Machine-Learning/Neural-Networks
- Biology/Pharmacology/Drug-Discovery
Reasoning: Primary tag reflects the technical method; secondary reflects application domain. If note focuses more on biological insights than ML technique, reverse the priority.
Ask: How established is this concept?
| Maturity Level | Tag Strategy |
|---|---|
| Canonical (in textbooks for 20+ years) | Use standard academic hierarchy |
| Established (widespread in journals/practice) | Follow field conventions |
| Emerging (active research, no consensus) | Use broader tags, avoid premature specificity |
| Speculative (blog posts, tweets, hype) | Tag the underlying established concepts |
Example:
Note: "GPT-4 Prompt Injection Attacks"
Maturity assessment:
- GPT-4: Very new (2023)
- Prompt Engineering: Emerging (2020s)
- Security vulnerabilities: Established
Decision: Tag using established concepts, not bleeding-edge labels
Conservative tag:
Computer-Science/Artificial-Intelligence/Natural-Language-Processing/Security
Alternative (if focusing on prompt engineering):
Computer-Science/Artificial-Intelligence/Prompt-Engineering
Reasoning: "Prompt injection" is too new and unstable as terminology. Anchor in established security or NLP concepts, then add emergent layer if needed.
Scenario: Note discusses a broad concept that could be tagged at multiple specificity levels.
Example: "Introduction to Machine Learning"
Options:
# Option A: Broad (appropriate for survey/intro)