taxonomic-interest-classification

Classify text corpora against hierarchical taxonomies to measure interest diversity and profile breadth.

13|2|Updated Feb 10, 2026
One-click install
npx skills add https://github.com/aaddrick/written-voice-replication --skill taxonomic-interest-classification-aaddrick
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: taxonomic-interest-classification
Source: https://github.com/aaddrick/written-voice-replication/tree/main/.claude/skills/taxonomic-interest-classification
Command: npx skills add https://github.com/aaddrick/written-voice-replication --skill taxonomic-interest-classification-aaddrick

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It turns a raw collection of posts, comments, or articles into a measurable interest profile, replacing subjective impressions of someone's interests with quantified diversity metrics and a defensible polymathic-versus-specialist classification. ## Core Features & Use Cases - Taxonomy Mapping: Maps corpus items to established hierarchical taxonomies such as the IAB Content Taxonomy using metadata, LLM-assisted, keyword, or hybrid approaches. - Diversity Metrics: Computes Shannon entropy, Simpson's diversity index, and a composite orthogonality score with a defined interpretation scale. - Profile Classification: Classifies profiles as Deep Specialist through Polymathic, identifies interest pillars, and cross-checks scores against distribution shape. - Use Case: Given a Reddit data export of 400 posts and comments, map each item to IAB Tier-1 categories, compute an orthogonality score, and produce a report identifying the user's primary interest pillars for downstream archetype assignment. ## Quick Start Analyze my writing corpus in this project and classify my interests using the taxonomic interest classification workflow, writing the report to docs/analysis.

Frequently Asked Questions about taxonomic-interest-classification

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I measure interest diversity from a text corpus?

Map each corpus item to a hierarchical taxonomy such as the IAB Content Taxonomy, build a category distribution table, then compute Shannon entropy, Simpson's diversity, and an orthogonality score. The orthogonality score classifies the profile from Deep Specialist to Polymathic.

What taxonomy should I use for content classification?

The IAB Content Taxonomy v2.0+ is the recommended baseline, offering 26 Tier-1 categories and 366+ Tier-2 subcategories. Adapt it when the corpus is domain-specific or when more than 15% of items fall into Uncategorized.

What is the difference between Shannon entropy and Simpson's diversity index?

Shannon entropy measures the information content of a distribution and is sensitive to rare categories, while Simpson's diversity measures the probability that two random items come from different categories. Using both together captures evenness and concentration resistance.

How many items do I need for meaningful diversity analysis?

Corpora under 30 items are too small for diversity metrics; only raw category counts should be reported. Corpora of 30-100 items can be analyzed with low-confidence caveats, and over 100 items supports full analysis.

When should I not use taxonomic interest classification?

Avoid it for single-document topic extraction, where NMF or LDA topic modeling fits better, and for keyword frequency analysis without categorical framing. It measures what a distribution is, not what content should be recommended.