Intelligent Document Classification is an web-based document classification application deployed on the [DAIR Cloud](https://www.canarie.ca/cloud). Using Optical Character Recognition, Natural Language Processing, and Full-Text Search, the Automated Document Classification and Discovery BoosterPack can automate the creation of document metadata.

Architecture

Features

✅ Optical character recognition using Tesseract

✅ Natural language processing using PyTorch and Hugging Face Machine Learning Datasets

✅ Fulltext search and metadata storage using Elasticsearch

✅ Web interface for document upload and document search using ReactJS

✅ Event Streaming using Apache Kafka

Run Local

# Build application
docker-compose -f docker-compose-dev.yml build

# Run application
docker-compose -f docker-compose-dev.yml up -d

License

Intelligent Document Classifcation is available under the Apache License V2.