Extract data from your documents with
unrivaled accuracy
Datacie licenses parts of its data extraction technology to help developpers working with documents save time and gain in granularity.
Extract data from your documents with
Document Structuration
Turn your documents into clean and structured data sources
Analyzing documents automatically is complex and error-prone. Leverage our document structuration API to make your documents free from any data preprocessing.
Usage-based Pricing
Pay for what you need only. Pricing starts from $9ct per page and is gradually decreasing every 1000 pages.
Layout Detection Comprised
In each page, the regions of interest (blocks) are identified and categorized into layout elements, allowing you to filter noisy components easily.
Cloud-API or On Premises-Docker
Host Datacie models on your own servers making sure that no sensitive data leaves your network.
Reading Order Retrieval
Once the physical structure of a page is built, our models detect the reading order between the components of the pages’ logical structure.
Built-in Optical Character Recognition
Scanned images and documents that contain non-standard codepoint encoding are detected automatically and sent to OCR systems.
Multiple Language & Format Support
Our models currently support 15 languages and accept a wide variety of input formats (A series, 4:3, 16:9, etc.).
Free Data Extraction Audit
Book a meeting with an Artificial Intelligence expert and learn more about how to automate your manual document processing.
Tabular Data Extraction
Extract data tables from your documents with certainty
The days of manually identifying, extracting and auditing data tables are long gone. While many offer this service, our level of precision on complex use-cases is still unmatched.
Usage-based Pricing
Pay for what you need only. Pricing starts from $5ct per page and is gradually decreasing every 1000 pages.
Handle Complex Table Layout
Do your documents contains tables with spanning cells, multi-level headers and heading rows? Not a problem for our technology.
Cloud-API or On Premises-Docker
Host Datacie models on your own servers making sure that no sensitive data leaves your network.
Metadata Augmentation
Our algorithms can augment data cells with contextual information such as numerals, units and footnotes details.
Built-in Optical Character Recognition
Scanned images and documents that contain non-standard codepoint encoding are detected automatically and sent to OCR systems.
Matrix Table Reconstruction
Translating pivot tables into their flattened representation to simplify storage and ensure bias-free analyses.
At Datacie, our expertise comes from our passion
Our team is composed of technology enthusiasts and business leaders that all believe in a world where people invest their time in what bring them value rather than in performing repetitive tasks.
We would love to help you achieve your goals.
Datacie SA
EPFL Innovation Park - Le Garage
Chemin de la Raye 13
CH-1024 Ecublens
About us
Datacie is an information company specializing in Document Intelligence that provides unique, actionable and fully customizable datasets to financial institutions, global corporations and non-profit organizations. Our mission is to empower research and decision-making by democratizing access to high-quality, professional-graded datasets directly built from the world's most trusted content sources.
© 2023 Datacie Sàrl. All rights reserved.
Made in Switzerland.