Kaspersky Machine Learning for Anomaly Detection 4.0 has the following new capabilities and improvements:
Model builder: Functionality has been added that allows you to create, modify, and delete your own ML models and elements of ML models based on a neural network and/or diagnostic rules.
Markups: Functionality has been added that allows you to create, modify, and delete markups for training and inferencing ML models.
Monitored asset hierarchical structure: Functionality has been added that allows you to define the way in which the assets of the monitored asset are organized in the form of a tree of primary and functional elements in Kaspersky MLAD. The primary elements of the hierarchical structure are represented by assets and tags, which can be managed in the Assets section. The format of the configuration file for downloading tags has been changed to XLSX. When the configuration file is downloaded, the assets of the hierarchical structure are also downloaded. The functional elements of the hierarchical structure are represented by ML models, ML model templates and markups. The monitored asset hierarchical structure is displayed in the Assets and Models sections, as well as in the Presets section when creating or modifying presets.
Roles: Functionality has been added that allows you to manage roles and select access rights to application functions for them. Role management is available in the Roles section.