With the Data Capture Configuration Tool, you can upload vendor data from an ERP system into OnBase and configure this data for use in your Intelligent Capture for AP solution. The data can be configured as you prepare it for upload; configure a Document Type and a primary Keyword Type for the vendor name/number, and associate the vendor name/number with uniquely identifying values that will physically appear on the documents (e.g., phone numbers, street addresses, email addresses, etc.). Then associate each identifier value with a regular expression. Optionally, you can also configure a secondary Keyword Type for the vendor name/number that complements the primary Keyword Type (e.g., If you configured a primary Keyword Type for the vendor number, you can configure a secondary Keyword Type for the vendor name, and vice versa.). Once you have finished configuring the vendor data, it can be uploaded into your OnBase system.
When vendor data is configured in this manner, the Intelligent Capture for AP engine first attempts to classify the document based on the default Document Type configured for the scan queue. If the Document Type cannot be determined at this point, the ICAP engine searches documents for the configured regular expressions. Once the regular expressions are found, the engine searches for identifier values that match the forms of these regular expressions. Then the engine determines the best layout match based on the page and sequence of the identifier values, and it assigns the associated Document Type and, if valid, the primary Keyword Value that is specified as the vendor name/number. Once the primary Keyword Value has been assigned, any matching secondary Keyword Value for the complementary vendor name/number is also assigned. Following this, the engine searches for any matching AutoFill Keyword Sets that have been configured for the Document Type or the Keyword Type. If a match is found, the remaining Keyword Values for the Document Type are populated and assigned Suspect Level scores of 100 so that they will not be marked suspect. If the secondary Keyword Value for the vendor name/number is assigned a different value by the AutoFill than it was assigned as the complementary value of the primary Keyword Value for the vendor name/number, the value assigned by the AutoFill replaces the original value. If no match is found among the configured AutoFill Keyword Sets, however, the engine indexes the remaining values in the normal fashion (i.e., according to training and/or manual configuration).
If an AutoFill Keyword Set match is found for an extracted PO Number value, depending on your configuration, this AutoFill Keyword Set is triggered to populate the vendor Keyword Values. If any conflicting values have been assigned according to the regular expression extraction method described above, the AutoFill values are given precedence and replace the extracted values that do not match.
Unlike the Secondary Keyword Values, the Primary Keyword Value for the AutoFill Keyword Set can be marked suspect since it was found by other (i.e., non-AutoFill) means.
As the Intelligent Capture for AP engine identifies vendor names/numbers on the documents in a batch, by default, it automatically separates the documents according to individual vendors. The engine then separates the documents belonging to each vendor according to individual PO and Non-PO invoice values. If no vendor has been identified on a page, the engine searches for an Invoice Number or a PO Number and, given sufficient Suspect Level scores, separates the page as a new document. This process allows for convenient, upfront document separation for single- and multi-page invoices.