hierarchical-taxonomy-clustering — Claude Code skill: install
The hierarchical-taxonomy-clustering skill combines similar category paths from multiple sources into a shared hierarchy. Use it to standardize product categories for cross-source analysis.
Apache-2.0 1 installs ★ 1,933 updated 2026-10-24
npx skills add benchflow-ai/skillsbench --skill hierarchical-taxonomy-clustering -a claude-codeWhat hierarchical-taxonomy-clustering does
The skill takes category paths from several sources, cleans and normalizes their labels, then groups related paths. Its output adds unified taxonomy levels to a DataFrame and exports one CSV with the records and another with the category structure.
The workflow has four stages: prepare the data, create weighted embeddings, cluster recursively, and export the results. It relies on the Python packages pandas, numpy, scipy, sentence-transformers, nltk, and tqdm. The setup instructions also download NLTK’s wordnet and omw-1.4 datasets, which the lemmatization step uses.
Run the full workflow with scripts/pipeline.py. Category labels come from weighted word frequency and lemmatization; clustering uses cosine distance at each level. The documented cluster counts are guidance, not fixed requirements for every taxonomy.
What it helps with
- merge similar category paths from separate sources
- create a shared hierarchy from inconsistent category labels
- export categorized records and a taxonomy structure as CSV
- align product categories for analysis across platforms
Example prompts for the agent
- Combine category paths from several sources into one taxonomy.
- Prepare the data and run the full workflow with
scripts/pipeline.py. - Group the categories by level and export the resulting CSV files.
How to install hierarchical-taxonomy-clustering
1Skills CLI (Vercel)
npx skills add benchflow-ai/skillsbench --skill hierarchical-taxonomy-clustering -a claude-code2Manually: put the files in place
~/.claude/skills/hierarchical-taxonomy-clustering/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/hierarchical-taxonomy-clustering/
1Skills CLI (Vercel)
npx skills add benchflow-ai/skillsbench --skill hierarchical-taxonomy-clustering -a codex2Manually: put the files in place
.agents/skills/hierarchical-taxonomy-clustering/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add benchflow-ai/skillsbench --skill hierarchical-taxonomy-clustering -a cursor2Manually: put the files in place
.cursor/skills/hierarchical-taxonomy-clustering/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install benchflow-ai/skillsbench tasks-extra/taxonomy-tree-merge/environment/skills/hierarchical-taxonomy-clustering2Manually: put the files in place
.github/skills/hierarchical-taxonomy-clustering/SKILL.md1Skills CLI (Vercel)
npx skills add benchflow-ai/skillsbench --skill hierarchical-taxonomy-clustering -a gemini-cli2Platform CLI
gemini skills install https://github.com/benchflow-ai/skillsbench --path tasks-extra/taxonomy-tree-merge/environment/skills/hierarchical-taxonomy-clusteringManually: put the files in place
.devin/skills/hierarchical-taxonomy-clustering/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add benchflow-ai/skillsbench --skill hierarchical-taxonomy-clustering -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about hierarchical-taxonomy-clustering
How do I install hierarchical-taxonomy-clustering in Claude Code?
Run npx skills add benchflow-ai/skillsbench --skill hierarchical-taxonomy-clustering -a claude-code. The skill file is located at ~/.claude/skills/hierarchical-taxonomy-clustering/SKILL.md.
What does hierarchical-taxonomy-clustering do?
It groups related category paths from different sources into a shared multi-level taxonomy. The results can be exported as CSV files.
Which Python packages does the skill need?
Install pandas, numpy, scipy, sentence-transformers, nltk, and tqdm. The setup also downloads NLTK’s wordnet and omw-1.4 datasets.
Source text
SKILL.md · Apache-2.0
---
name: hierarchical-taxonomy-clustering
description: Build unified multi-level category taxonomy from hierarchical product category paths from any e-commerce companies using embedding-based recursive clustering with intelligent category naming via weighted word frequency analysis.
---
# Hierarchical Taxonomy Clustering
Create a unified multi-level taxonomy from hierarchical category paths by clustering similar paths and automatically generating meaningful category names.
## Problem
Given category paths from multiple sources (e.g., "electronics -> computers -> laptops"), create a unified taxonomy that groups similar paths across sources, generates meaningful category names, and produces a clean N-level hierarchy (typically 5 levels). The unified category taxonomy could be used to do analysis or metric tracking on products from different platform.
## Methodology
1. **Hierarchical Weighting**: Convert paths to embeddings with exponentially decaying weights (Level i gets weight 0.6^(i-1)) to signify the importance of category granularity
2. **Recursive Clustering**: Hierarchically cluster at each level (10-20 clusters at L1, 3-20 at L2-L5) using cosine distance
3. **Intelligent Naming**: Generate category names via weighted word frequency + lemmatization + bundle word logic
4. **Quality Control**: Exclude all ancestor words (parent, grandparent, etc.), avoid ancestor path duplicates, clean special characters
## Output
DataFrame with added columns:
- `unified_level_1`: Top-level category (e.g., "electronic | device")
- `unified_level_2`: Second-level category (e.g., "computer | laptop")
- `unified_level_3` through `unified_level_N`: Deeper levels
Category names use ` | ` separator, max 5 words, covering 70%+ of records in each cluster.
## Installation
```bash
pip install pandas numpy scipy sentence-transformers nltk tqdm
python -c "import nltk; nltk.download('wordnet'); nltk.download('omw-1.4')"
```
## 4-Step Pipeline
### Step 1: Load, Standardize, Filter and Merge (`step1_preprocessing_and_merge.py`)
- **Input**: List of (DataFrame, source_name) tuples, each of the with `category_path` column
- **Process**: Per-source deduplication, text cleaning (remove &/,/'/-/quotes,'and' or "&", "," and so on, lemmatize words as nouns), normalize delimiter to ` > `, depth filtering, prefix removal, then merge all sources. source_level should reflect the processed version of the source level name
- **Output**: Merged DataFrame with `category_path`, `source`, `depth`, `source_level_1` through `source_level_N`
### Step 2: Weighted Embeddings (`step2_weighted_embedding_generation.py`)
- **Input**: DataFrame from Step 1
- **Output**: Numpy embedding matrix (n_records × 384)
- Weights: L1=1.0, L2=0.6, L3=0.36, L4=0.216, L5=0.1296 (exponential decay 0.6^(n-1))
- **Performance**: For ~10,000 records, expect 2-5 minutes. Progress bar will show encoding status.
### Step 3: Recursive Clustering (`step3_recursive_clustering_naming.py`)
- **Input**: DataFrame + embeddings from Step 2
- **Output**: Assignments dict {index → {level_1: ..., level_5: ...}}
- Average linkage + cosine distance, 10-20 clusters at L1, 3-20 at L2-L5
- Word-based naming: weighted frequency + lemmatization + coverage ≥70%
- **Performance**: For ~10,000 records, expect 1-3 minutes for hierarchical clustering and naming. Be patient - the system is working through recursive levels.
### Step 4: Export Results (`step4_result_assignments.py`)
- **Input**: DataFrame + assignments from Step 3
- **Output**:
- `unified_taxonomy_full.csv` - all records with unified categories
- `unified_taxonomy_hierarchy.csv` - unique taxonomy structure
## Usage
**Use `scripts/pipeline.py` to run the complete 4-step workflow.**
See `scripts/pipeline.py` for:
- Complete implementation of all 4 steps
- Example code for processing multiple sources
- Command-line interface
- Individual step usage (for advanced control)
© benchflow-ai · Apache-2.0 · benchflow-ai/skillsbench