Role required
To Train a model you need to be assigned to either the Application Administrator or Records Manager role in Records365.

Feature Requirement
Classification Intelligence is delivered through the use of machine learning techniques leveraging the binary content of records stored in Records365. This means in order to use the Classification Intelligence add-on, Records365 must be managing the binary content of at least one content source.


The training process leverages records that already have a business classification in Records365 in order to build a model that can, with high confidence, predict a record category (business classification) for future content.

Currently content in Records365 can be assigned a record category (business classification) in the following ways:

  • Metadata based, through the use of the automated Rules Engine,
  • Manual, through the manual assignment of record categories and the Reschedule feature.
  • Content based, through the use of the Classification Intelligence add-on.

Training a category

Some areas of a file plan may be better suited to machine learning, and so one or more record categories can be targeted for inclusion in a model through the training process. To build a model with a selected set of categories perform the following:

  1. In the left-hand navigation menu, click on the Manage section and select File Plan.
  2. Select two or more record categories from the file plan. Note selecting a parent node will automatically select all of the child nodes. Some things to consider when selecting record categories to train on are:

    • Two or more categories must be selected
    • For a category to be eligible for inclusion into the model there must be at 50 existing records assigned that category. At most, the latest 500 documents for each category will be used.
  3. Click the Train button.

The Train button will be unavailable in the following scenarios:

  • an insufficient number of record categories have been selected
  • a training experiment is currently in progress
  • a training experiment has already been performed in the last 24hrs