How to perform a fraud, quality or classification analysis on a document?

To perform a fraud analysis, a quality analysis or a classification of a document file loaded in Athento, there are two complementary operations:

op_resistant_analyze_document.py
op_resistant_check_document_analysis.py

It should be noted that this analysis is performed through an integration with the Resistant AI application, so for it to work correctly, the RESISTANT_CLIENT_ID and RESISTANT_CLIENT_SECRET settings must be configured.

The first thing to do is to install both operations through the marketplace. Once this is done, you can configure the op_resistant_analyze_document operation to indicate the type of analysis to be performed. The options specified for this parameter indicate the way in which the analysis will be performed internally:

Once this configuration is done, you can proceed to execute or automate the execution of these operations.

When you run op_resistant_analyze_document on a document, you can see that the document characteristics indicate both the type of analysis that was sent to be performed, as well as an identifier of the analysis request generated by Resistant. These data are necessary to obtain the results of the analysis.

After this, we will have to execute the following operation op_resistant_check_document_analysis on that document. In this way we will be able to obtain the results of the analysis. Depending on the type of analysis executed, some or other characteristics will be created:

It is possible that at the moment of executing this operation the results are not yet ready, since the analysis normally takes a few seconds. In such a case, the same feature will indicate that the results are not yet ready.