Intelligent Data Extraction
Most organisations receive documents that they need to re-key into other systems for processing or further analysis. These could be invoices, credit notes, bank statements or credit card statements. These documents usually contain tabular or forms based data that repeats over many pages.
Some organisations need to go a step further. They also need to extract information from the unstructured text contained within the table or form.
How can Symilarity help?
Symilarity’s Intelligent Extraction system processes multi-page PDF documents and extracts and normalises data for use in other downstream processes.
Document extractions can be configured through a user console, with no programming requirement. Cleansing and normalisation rules can also be applied to the extracted data, to ensure it is in the correct format for the downstream process.
Configured applications can be dedicated to one type of document or handle many types. The application is deployed as an API (Application Programming Interface) which enables it to be integrated into custom processes, such as robotic process automation.
Symilarity provides the user console and API endpoint as a single tenanted service, not shared with any other clients.
For confidentiality reasons, documents or processed output are not retained.
Symilarity provides its service on a dedicated Virtual Machine. Service pricing is based upon the size of the machine. Usage is not limited to a number of pages or documents, only by the capacity of the machine.
Client Cloud/On Premise
Where document content is confidential and clients do not wish to transmit them to external APIs, Symilarity can provide the API on customer-managed infrastructure, inside the customer’s firewall.