Document validation concept
Intelligent Document Processing is a quality driven process which is controlled by elDoc system by analyzing confidence levels (per-character) of the retrieved data, e.g.: documents scanned with a good quality provide results with a high confidence levels whereas documents of the low scanned quality or with artifacts like remarks over printed text may provide results with low confidence levels.
Document Validation is required in the following cases:
- Some required field/data value in document was recognized with confidence level lower than confidence threshold defined in the recognition rules;
- Proper recognition form (RecoForm) was not found (not matched) for the document;
- Two or more recognition forms (RecoForms) were matched with the same confidence level for the document;
- Errors occurred during Conversion phase.
Performing document validation
- Documents requiring validation are moved by the elDoc BPM engine to the Inbox view of the assigned users (validators);
- Additionally assigned user(s) receive notification via email with summary on the documents which are pending validation with direct links to them;
- elDoc system allows only one single user to edit document on the Validation phase, for other users document will be locked for edits and available for review only;
- Validation assumes that user must verify recognized and retrieved data from the document and correct it where required;
- elDoc system will highlight characters retrieved with confidence level below defined threshold levels on the RecoForm with red color, and show the minimal confidence level in the box on the right to the respective field;
- Once visual check (validation) is completed, the document can be sent for further processing by pressing the Validate button;
- Cancel button will cancel document further processing;
- In case the document returns back for validation from Document conversion stage, the user (Validator) must make sure that the recognition data is correctly specified, make appropriate edits if required and send the document for re-processing by clicking on Validate button.
Multi-page multi-document files validation
PDF (or TIFF) files which contain several (multi-page) documents (so called multi-page multi-document files) provide controls for performing validation of individual documents inside the given file.
- Select document which requires validation (marked with [Requires validation]) in the Recognition data area:
- Once required document selected elDoc system provides set of controls for performing validation:
- #1 - Switch between documents inside multi-document file;
- #2 - Edit current recognition data for the selected document inside multi-document file;
- #3 - Delete current recognition data and document identified inside multi-document file;
- #4 - Add additional recognition data;
- #5 - Page range of the document inside multi-document file;
- #6 - Recognition form name of the currently selected document inside multi-document file;
- Use provided controls to perform corrections of the retrieved data and complete validation of the selected document inside multi-document file;
confirms correctness of recognized fields for the currently selected (in the Recognition data area) document and switches to the next document which requires validation.
Recognition document will be moved to the next step according to the defined process (e.g.: conversion to the target document after recognition, Repository or Archive) only after all documents inside multi-document file are validated;
|cancels document and moves it directly to Archive (with Status = Cancelled)|
|resends document to recognition|
|saves currently selected (in the Recognition data area) document|
|downloads file content as text document|
|navigates to the previous document (in the current view)|
|navigates to the next document (in the current view)|