Like in other industries, banking industry do have many processes where human intervention is involved. Most of these processes start with document processing. In banking industry, processes which are document-driven, could be digitalized.
Several banking documents are of varied file types. To list, they could be of – form data, image file, text file, PDF’s and much more. One of the use cases around this was, from one of our banking clients, to capture information from the running document. The machine learning capability of the Datacap Accelerator helps in extracting relevant information from these documents.
Datacap Accelerator provides support to various file formats for processing. Some of them - tiff, jpeg, pdf, doc.. Few could be running text like the letter of credit or even swift documents.
These swift documents are elaborate document that’s generated when a transaction is completed upon transfer to a bank. It’s more like a confirmation of payment made from the bank and informs the beneficiary of the details of the transaction. Since these are running text – most of it, is in the form of block of words like a paragraph. Datacap Accelerator helps in extracting specific information from these block of text.
For instance –
* To extract the amount from description.
* To extract relevant keywords like the city name or country name from an address field.
Furthermore, this capability of IBM Datacap Accelerator helps in better extraction of information from various types of documents.
Try out different capabilities of IBM Datacap Accelerator by visiting this site.