21 Mar Shamrock’s Intelligent Data Extraction
Let’s say your company has a million paper invoices and credit memo that needs to be collected into some type of database for accessibility and storage. Without Shamrock’s intelligent data extraction, this might entail hiring a few people to read through all of these invoices and credit memos, determine whether the document is an invoice or credit memo, and manually type the document’s information into a database. This means:
- Time is lost because of the time it takes for someone to read all of the invoices and credit memos
- Data is lost or inaccurate because human errors are prone to happen
- Money is spent on hiring so many people to read the documents and type it into the database to make up for the time lost
Instead, use an intelligent data extraction software. Shamrock has a document capture and data extraction software that uses cloud technology and advanced machine learning to take physical and electronic documents and extract data to pass into ERPs, database, or other systems.
So instead of having a few people read through all of documents, these paper invoices and credit memos can be scanned and directly captured into Shamrock’s intelligent data extraction software. Other sources include email server, directory system, and other sources.
The intelligent data extraction software then automatically classify the documents into invoices or credit memo. More document types can be added to the system for the software to classify.
After the invoices and credit memos are sorted out, data is extracted from the documents using advanced OCR (optical character recognition) algorithm. This algorithm can read typed fonts as well as handwritten fonts. With this, we can find all of the invoice numbers, invoice dates, invoice amount, and other fields from all of the documents. Our intelligent data extraction software can also retrieve invoice line items such as item description, quantity, and item amount.
All of the invoice and credit memo extracted data can be validated. For example, if we want all invoice date to be in a format of MM/DD/YYYY, this step will ensure that the invoice date is in a format of MM/DD/YYYY. If the invoice date is not in the correct format, it’ll kick the document batch into validation mode where a user logs in and validate/edit the data that the software extracted.
At the very end of the process, the invoice and credit memo data gets exported into the next system. This includes all of the data extracted from the previous steps as well as the scanned in invoices and credit memos that was captured in the first place. Data can be exported as index files, through a web service, and other ways.
Shamrock’s intelligent data extraction software has a nice and easy to use interface. The time it takes for people to read the invoices and credit memos and type into a database is removed and the validation reduces the chance for human errors thus making this solution time efficient and accurate which in turn will be cost-efficient.
Written by Katherine Wu