Glossary

Data Source - a connection to a certain data storage (files / database tables).

Data - a dataset with saved configured metadata (connection source, description, labeling, etc.).
Examples of data objects: one csv file, one SQL database table.

Data Labeling - a description of the properties of the columns of the registered data object in the system.

Metric - an executable test (script) performed on the data, producing a result in the form of a graph and/or a scalar value.

Model - a predictive analytical model.

Transformation - an executable script performed on the data, producing a result in the form of a dataset with a number of columns greater than or equal to the number of columns in the original dataset. The application of a model to a dataset containing a set of features is defined in the system as a transformation that adds a column with predictions to the original dataset.

Project - a monitoring project. It consists of a dataset, metrics, and (optionally) models with a user-configured execution mode.

Regulation - the schedule for running the monitoring project.

Dashboard - a section of the project with the results of metric calculations.

User Dashboard (report) - a collection of text panels and graphs created by the user based on the results of the monitoring project calculations. One user report can include results from several monitoring projects.

Project Template - a pre-configured saved set of metrics, their parameters, and (optionally) regulation settings.