ZoteroAutoTagger

ZoteroAutoTagger is a Python package that uses topic modeling and natural language processing (NLP) to automatically tag items in a Zotero library. It helps researchers organize their libraries by content-based topics without manual tagging.

---

Features

• Extracts topics from paper abstracts and full-text using BERTopic
• Automatically applies human-readable topic labels
• Integrates with Zotero via its Web API to tag items
• Uses YAML config files for customization and reproducibility
• Works in batch — no manual tagging needed

---

Installation

git clone https://github.com/nick-rodriguez/ZoteroAutoTagger.git
cd ZoteroAutoTagger
pip install -e .

---

Input and Configuration

To generate the library.json input file, export your Zotero library or a collection:

1. Open Zotero.
2. Right-click on the collection (or your full library) you want to analyze.
3. Choose Export Collection (or Export Library).
4. Select Format: Zotero RDF or Format: JSON depending on your setup.
   • This tool expects a JSON export.
5. Uncheck all optional checkboxes unless you want file attachments.
6. Save the file as library.json in your project directory or update the input_file path in your config.

This package requires two YAML configuration files:

config.yaml

input_file: "library.json"
output_topics_file: "topics_output.json"
output_docs_file: "docs_output.json"
topic_labels_yaml: "topic_labels.yaml"

zotero_config.yaml

api_key: "YOUR_ZOTERO_API_KEY"
user_id: "YOUR_USER_ID"
library_type: "user"  # or "group"

Optionally override topic names with topic_labels.yaml.

---

Usage

Run the full pipeline with:

zoteroautotagger

This will:

1. Generate topics with BERTopic
2. Apply readable labels (optional)
3. Push tags to Zotero

Make sure your YAML files are populated and renamed (remove .example).

Note: This will create new tags in your Zotero library. If you want to do a dry run and inspect the results manually before pushing to Zotero, then run the tagging and topic modeling modules separately.

---

📝 License

MIT License