Integrating machine learning models in multidimensional applications

A computer-implemented method performed by a computing device including at least one processor, the method comprising:

• generating, on a display device, a user interface configured to integrate a machine learning (ML) model into a target application, wherein the user interface provides a list of available ML models for selection on the display device;
• in response to a selected ML model being selected from the list of available ML models, retrieving a definition file that represents the selected ML model;
• parsing the definition file to (i) identify one or more input features defined for the ML model, and (ii) identify one or more output predictions that the ML model is configured to generate;
• mapping, via the user interface, the one or more input features of the ML model to one or more input data locations from the target application, wherein the one more input data locations provide input values to the one or more input features;
• mapping, via the user interface, the one or more output predictions of the ML model to one or more result locations associated with the target application;
• generating and executing, by at least the computing device, an executable rule including executable code that at least (i) defines the mappings between the ML model and the target application, and (ii) transmits data values from the one or more input data locations to the ML model as input;
• executing, by at least the computing device, the ML model based on the input to generate the one or more output predictions;
• and
• inserting the one or more output predictions into the one or more result locations associated with the target application.

A non-transitory computer readable medium including instructions stored thereon that when executed by at least the processor cause the processor to:

• generate a user interface configured to integrate a machine learning (ML) model into a target application;
• parse a definition file that represents the selected ML model to (1) identify one or more input features defined for the ML model, and (2) identify one or more output predictions that the ML model is configured to generate;
• map the one or more input features of the ML model to one or more input data locations from the target application, wherein the one more input data locations provide input values to the one or more input features;
• map the one or more output predictions of the ML model to one or more result locations within the target application;
• generate executable script configured to function as a data controller and data communication interface to obtain the input values from the one or more input data locations and pass data between the target application and the ML model;
• wherein the executable script is configured to execute the ML model with the input values and obtain the one or more output predictions from the ML model;
• and
• in response to obtaining the one or more output predictions, based on the executable script, insert the one or more output predictions in the one or more result locations within the target application.