Last Updated on

What’s New In Transact 4.5?


Machine Learning | Support for Multiple JSON Files

 

Whenever you perform machine learning for a document, a new machine-learning-extraction subfolder is created in the Batch Class folder (SharedFolders/<Batch Class>) on the Ephesoft Transact server. This subfolder contains JSON files with machine learning data for each Document Type and Index Field.

C:\Users\Ephesoft\AppData\Local\Microsoft\Windows\INetCache\Content.Word\18.png

 

If, for example, another user learns the same Index Field under the same Document Type and Batch Class via a Web Service, the application will save the JSON file for that learning as well.

C:\Users\Ephesoft\AppData\Local\Microsoft\Windows\INetCache\Content.Word\19.png

 

Now, the next time you extract data from the document, the system will compare the anchors around the extracted value with anchors saved in all existing JSON files. The value with the highest confidence will be shown as the extraction result. The JSON files are then merged into one file containing the latest learning information.

Note: Anchors are basically words surrounding a specific value. During extraction, anchors help to determine if expected keywords are found, and if the value’s neighbors match any of the neighbors found during training.

This feature helps ensure that all machine learning results are saved and used during further data extraction.

Please keep in mind that manual editing of JSON files is not recommended.