You can train classification models to recognize when documents are a certain type of document. To train a model, upload files that are good representations of the type of documents that you want to classify. For example, to create a classification model that can identify legal documents, upload a set of typical legal documents.
To improve the accuracy of your classification models, you can continue to train the models by uploading additional documents.
After the classification model is trained, add the classifier to a data classification plan. The classification information is available for Sensitive Data Governance data sources that are associated with the data classification plan.
If you have an Azure account, you can also use the predefined Azure classification models.
Support
-
Classification is available for Sensitive Data Governance data sources.
-
Files in the following languages are supported:
-
To use the Commvault classification models, the files must be in English.
-
To view the languages supported for the Azure classification models, go to Language support for Form Recognizer on the Microsoft documentation website.
-
-
The Azure classification models have a 50 MB file size limit and analyze image files and PDF files. These limits are imposed by Microsoft.