Evolution AI uses computer vision and natural language processing (NLP) to extract data from PDFs, scans and machine-readable text. Our platform is pre-trained on 25 million documents, so it can learn to read new documents without substantial additional training.
Our Evolution Catalyze managed-service leverages Transcribe and NLP to perform data extraction tasks. All setup, AI training and QA is performed by our expert AI consultants. Our pool of human-in-the-loop annotators provide human oversight and verification. High quality data can be provided in JSON or CSV, via FTP or our REST API.
All features from Evolution Transcribe & NLP
99.9% accuracy from first day of deployment
A fully managed service
Unlike traditional OCR, our intelligent data extraction engine does not need any templates or rules. Evolution Transcribe can find information by itself, because it inherently understands the way documents are put together.
Evolution NLP handles "electronic" text, like that found in websites or emails. You can use it to go one step further in your data-entry task. For example, to classify invoice line-items after extracting them with Evolution Transcribe.
Evolution AI offer a complete suite of tools for any data extraction task. Choose our Catalyze service if you're looking for a quick way to extract clean, perfectly structured data from thousands or even millions of documents.
Artificial intelligence means our products are self-learning. No explicit configuration is required to learn how to extract information from a new document. Just point and click—our products learn by themselves.
Very little training data is needed to achieve production-level accuracy. Active learning algorithms focus on the data that is most valuable, so AI training can be completed in a day or two.
Our neural-networks have a head-start when learning a new document. Our AI scientists have already pre-trained our models on 25 million documents from our proprietary document store.
Custom robustness checks and anomaly detection make sure your data quality is always maintained.
Evolution AI was funded by one of the largest AI R&D grants ever awarded by Innovate UK, the UK funding body for innovation.
A full audit trail is stored so the provenance of each data point can be tracked throughout its life cycle.
Catch any data quality problems before they cause downstream issues.
ISO/IEC 27001 is the international standard on how to manage information security. Evolution AI is certified ISO 27001 compliant, ensuring all information assets are held securely.
Yes, our AI models provide accurate confidence score that represent the probability of an error being found in each datapoint.
Yes, our Evolution Transcribe product read and understand handwriting. Depending on the legibility of the handwriting, lower accuracy should be expected compared to printed text.
Our language module add-on supports 40 European languages, Chinese, Japanese and Arabic.
Our comprehensive monitoring suite catches any data quality problems before they cause downstream issues. Our QA workflow is completely configurable, allow you to add as much human oversight as you wish. Automated anomaly detection and delta checking add an extra layer of safety.
We provide a fully documented REST API with code snippets in python, ruby, C# and javascript. We also supply code examples for RPA systems including UiPath, Automation Anywhere and Blue Prism.
We offer fully featured data extraction tools, including check-boxes and complex multi-page tables.
Our AI-powered interface makes it easy to extract content from complex tables and many other document formats.
We offer specialised functions for financial statements, including currency and unit detection.
Data Schema and data normalization are configured at the outset of the project so that downstream tasks can depend on consistent formatting of output data.
