# Automated PDF extraction using IDP

> The published case reports 93% accuracy for entity extraction and 95% for detecting relationships between entities.

URL: https://preview2.snowfox-ai.com/cases/pdf-extraction-idp
Last updated: 2026-11-20

An automatic document-processing pipeline combining intelligent document processing with supervised entity extraction from natural-language text.

- Sector: Financial services
- Built with: NLP
- Methods and stack: IDP, Supervised NER, Relationship extraction
- Entity extraction accuracy: **93%**
- Relationship detection accuracy: **95%**

## Challenge

A financial services company needed to make customer onboarding faster while accurately reading the entities and relationships in complex contracts.

## Scope

Read PDF contracts, identify the parties and powers granted, and make legal text easier to work with in the registration process.

## Approach

An intelligent document-processing pipeline combined document recognition with supervised extraction of entities and relationships from natural-language text.

## Outcome

The published case reports 93% accuracy for entity extraction and 95% for detecting relationships between entities.

Original case on snowfox-ai.com: https://snowfox-ai.com/cases/setor-financeiro/
