Create an extraction golden files benchmark set
If you want to create a set of golden files for extraction benchmarks, you can do this in one of the following ways:
-
Validate a set of documents using Project Builder and Test Validation.
-
Process a set of documents through Kofax Capture, Kofax Transformation - Server, Validation, and Knowledge Base Learning Server and save documents for online learning.
The following steps guide you through the steps to create your golden files using Project Builder and Test Validation.
You can create an extraction golden files document set by following these steps:
- Open the Documents window if it is not already open.
-
Add a new
document set that includes the documents you want to include in your golden files, ensuring that you provide a descriptive name for
the document subset such as
Extraction Golden Files.
A new document set is added to the Documents window and expands so you can see its document subsets automatically.
-
Convert your new document set to a benchmark set.
The document set type changes so that you can assign a classification for each document in the document set.
-
If a different view is in use,
switch to the
List View
.
-
Assign a
class for all documents in your benchmark set.
Each document is assigned a class for comparison after classification.
-
Click in the
Extraction Golden Files
document subset you created earlier and click Select All
from the toolbar or press Ctrl + A.
All documents in the selected document subset are selected in the List view.
-
Optionally, if your documents already have recognition data, right-click the selected documents, and then click
Clear Classification & Extraction Data
.
If you are not sure whether your documents have recognition data or not, perform this step anyway.
This removes any existing classification or extraction data and provides you with a clean state so you can build your golden files.
-
Press
F4 to perform recognition for the selected documents using the default
recognition engine.
If you want to select a different recognition engine, right-click the selected
Extraction Golden Files
documents, click Recognize
on the shortcut menu, and then select the desired recognition engine from the submenu.
A progress bar is displayed showing the recognition progress.
-
On the document shortcut menu, click
Classify
.
Each document in the benchmark set is assigned a classification result based on your project settings and training documents if available.
-
Once again, click
Select All
for your
Extraction Golden Files
document subset, and then click Save All
.
Tip To ensure you do not lose any changes, save your documents regularly.All changes to the documents are saved.
- Compare the classification result for each document against its assigned class. If any do not match, confirm that the assigned class is correct. If the assigned class is correct, then add the mismatched to the Classification Set and repeat steps 6 - 10 until the classification result matches the assigned class for all documents in your benchmark set.
-
If you added any documents to your
Classification Set, train your project. From the Ribbon
Process tab, in the
Train group, select
Separation & Classification
.
A progress bar is displayed showing the training progress.
-
Now that your classification results are correct, you can work on your extraction results. Once again, click
Select All
for your
Extraction Golden Files
document subset.This selects all documents in your subset.
-
Right-click the selected documents, and then click
Extract
from the shortcut menu.
Alternatively, you can skip this step and manually classify and extract the documents in Test Validation.
A progress bar is displayed showing the extraction progress.
-
You can view your extraction results by opening the
Extraction Results window.
If you have five or fewer fields on your documents, you can also add field columns to the List View in the Documents window. This does not however, give you any information about the confidence or alternatives.
The confidence of each extraction result is displayed in the Extraction Results window so that you see what needs improvement.
- If the extraction results for any document are incorrect, add that document to the Extraction Set.
-
Since your
Extraction Set is modified, retrain your project. From the Ribbon
Process tab, in the
Train group, select
Extraction
.
A progress bar is displayed showing the training progress.
-
Repeat steps 13 - 17 until you are happy your extraction results and you are ready to validate your documents.
Note Even if a document is in the Extraction Set, some fields can still be missed. This is not a problem because can fix them in Test Validation.
-
On the Ribbon
Process tab, in the
Test group, select
Validate
.
If your project has more than one validation step configured, you can select the desired step from the submenu.
The Test Validation window opens and is similar to the Kofax Transformation -Validation module.
- Process each document by validating each field that was not extracted correctly.
- On the Batch menu in Test Validation, select Exit to return to the main Project Builder window.
-
Once again, click
Select All
for your
Extraction Golden Files
document subset, and then click Save All.
-
On the Ribbon
Project tab, in the
File group, click
Save Project
.
Tip Create a backup copy of your processed golden files. This ensures that you can always access the backup copy if something happens to the working copy.Your project is saved and your golden files are now ready to perform Extraction benchmarks.