The platform can connect and read from Amazon S3, Google Drive, Box, DropBox and SFTP. You can also integrate with our API directly. The platform can ingest word documents, PDFs, images, and XML.
Our natural language processing techniques are primed with your industry-specific taxonomies and domain knowledge. Combining this with deep learning, gets us a much better contextual understanding of your data then was previously possible.
Once an understanding of your documents is established and key information is identified we can build models to automatically extract this information.
The extracted content is validated and those with low confidence scores are flagged for manual review. This allows the system to continually improve and ensures the system adapts to changes in your underlying data.
The platform can deliver the data back to you in structured formats such as XML, CSV, JSON or excel files directly to the original source (S3, Drive, Box, ...) or the data can be retrieved directly through our API.
Get in touch.