Machine-learning-based OKR generation

An application service provider server, comprising:

• a processor;
• an enterprise application that is utilized by an enterprise;
• a communication connection for connecting a remote computing system to the application service provider server via a network, wherein the remote computing system is operated by an enterprise user corresponding to the enterprise;
• and
• a computer-readable storage medium operatively coupled to the processor, the computer-readable storage medium comprising computer executable instructions that, when executed by the processor, cause the processor to:
• generate synthetic Objectives and Key Results (OKR) data using a generative pre-trained transformer bidirectional language model;
• self-supervise the generated synthetic OKR data based on desired OKR traits comprising at least one of rule-based criteria or model-based criteria;
• generate a set of positive synthetic OKR data labels with a gradient scale rating;
• synthesize a set of negative synthetic OKR data labels by self-supervising the positive synthetic OKR data labels;
• train a transformer-based, self-supervised bidirectional language model using the set of positive synthetic OKR data labels and the set of negative synthetic OKR data labels;
• cause execution of the enterprise application on the remote computing system;
• cause surfacing of a user interface on a display of the remote computing system during the execution of the enterprise application, wherein the user interface corresponds to a goalsetting feature of the enterprise application;
• receive, via the surfaced user interface, text input corresponding to an objective for an OKR;
• assess the text input based on the desired OKR traits using the trained transformer-based, self-supervised bidirectional language model;
• and
• cause surfacing of a user interface element on the user interface, wherein the user interface element comprises an indication of whether the text input satisfies each of the desired OKR traits.

A computer-readable storage medium having computer-executable instructions stored thereon, which, upon execution by a computer, cause the computer to perform operations including:

• generating synthetic OKR data using a generative pre-trained transformer bidirectional language model;
• self-supervising the generated synthetic OKR data based on desired OKR traits;
• generating a set of positive synthetic OKR data labels with a gradient scale rating;
• synthesizing a set of negative synthetic OKR data labels by self-supervising the positive synthetic OKR data labels;
• training a machine learning model using the set of positive synthetic OKR data labels and the set of negative synthetic OKR data labels;
• executing, via a network, an enterprise application on a remote computing system;
• causing surfacing of a user interface on a display of the remote computing system during the execution of the enterprise application, wherein the user interface corresponds to a goal-setting feature of the enterprise application;
• receiving, via the surfaced user interface, text input corresponding to an objective for an OKR;
• assessing the text input based on the desired OKR traits using the trained machine learning model;
• and
• causing surfacing of a user interface element on the user interface, wherein the user interface element comprises an indication of whether the text input satisfies each of the desired OKR traits.