Test Classification

This feature helps in understanding the results generated or the classifications done by DOCUMENT_ASEMBLER plugin. The user can input files for classification by dropping it or uploading through the Document type listing. The input tiff/pdf files could be single page or multi-page files. All the files will be processed according to the settings in the batch class. All the required plugins which are required for DOCUMENT_ASEMBLER plugin to execute for a particular configuration need to be present in the batch class workflow. By default the classification set in DOCUMENT_ASEMBLER plugin will be performed.

 

Existing Functionality – Version 4.0.5.0

There is no workflow based process for testing the classification results. The user has to manually select the Classification Type. The classification results also were not correct most of the time.

Click here for detailed information.

 

New Functionality- Version 4.0.6.0 and Above

A Workflow Switch is provided to test the Classification results for an input document before executing the batches.

If the workflow switch is set to ON, then ‘Classification Types’ list box will be disabled and the results will be generated as per the batch class configurations.

If the switch is set to OFF, then the results will be generated based on the manual configurations done in the ‘Classification Types’ list box on the UI.

F:\Enterprise\Product documentation 4060\images\release notes images\test classification.png

 

The classification results are shown in a tabular form with following details about the input images:

  1. Classification Type: The type of classification performed on the set of inputs.
  2. Document Type: the type of document classified for this input page.
  3. Document Identifier: The identifier of the document classified.
  4. Document Confidence: The confidence value for this document generated.
  5. Page Name: The name of the page given as input for classification.
  6. Page Identifier: The identifier of the page in accordance to the inputs provided.
  7. Page Classification: The classification done for the page. This contains the name of document type for which image is classified with the name of page type i.e. FIRST_PAGE, MIDDLE_PAGE or LAST_PAGE.
  8. Page Confidence: The confidence generated for this page after classification.
  9. Classification Sample: The name of the classification/learned file name from the “lucene-search-classification-sample” folders. This value is only populated in case of Search Classification. For all other classification it is set as “NA”.

Characteristics

  • This feature provides the results of the classification performed on the provided set of inputs.
  • The classification results which would be generated via batch processing can be viewed in a tabular form.
  • Using this functionality users can configure their batch class for optimum and best fit classification results. This classification can be executed for different settings and results can be compared.
  • The color code functionality highlights the document/page rows for which document confidence is lower than the threshold value of the classified document. These rows depict the documents which would require user interference in Review Validate modules.
  • The feature also provides us the functionality of download through which we can download the resulting batch xml structured output which contains details of all the details of processing.
  • Clear button functionality clears all the input files and the results generated from test-classification-folder.

 

Was this article helpful to you?

Engineering