Overview:
- Training statements follow the same process as invoice table annotation training.
How to Train Statements in Ocerra
Since statements are a new document type for the system, you will need to train them to ensure accurate data extraction. The training process involves the following steps:
1. Match Supplier Names
To begin, ensure that the supplier name on the statement matches the correct supplier. This step helps the system associate the statement with the appropriate supplier.
2. Annotate the Statement Table
In some cases, you may need to annotate the statement table to help Ocerra recognise and remember the layout. This process is identical to the table annotation process used for invoices.
Video lesson: Watch how to train the system with table-based annotation
A step-by-step guide
- Click the “Clean” button
- Click the “Table” button and move the cursor to the left. You will see a blue pointer on the left side of the screen
- Use this pointer to draw a new frame around the table:
- Click and release the cursor and start drawing the frame around the invoice table. When finished, release the cursor, and it will fixate the frame. Go to the next page if the table continues on the second and third pages.
- Click the “Annotate” button but don't close annotate window unless it is completed, so you can see the result.
Once this step is completed, Ocerra will remember the table layout and extract the data correctly for future statements from the same supplier.
Why Training Is Important
By matching supplier names and annotating statement tables, you ensure that Ocerra learns and adapts to the layout of your statements. This minimises manual corrections and enhances the efficiency of processing statements over time.
If you’re already familiar with table annotation for invoices, you’ll find the process for statements to be the same.
Follow the same steps to teach the system how to read and extract data accurately.