AI platform with automatic analysis data and methods for use therewith
AssigneeVirtuous AI, Inc.
InventorRory Donovan
Life of this patent
Estimating the term…
Abstract
A system operates by: generating, via a machine that includes at least one processor and a non-transitory machine-readable storage medium and utilizing a graphical user interface, a content analysis control panel; receiving, via the machine, content data; detecting, via one or more AI models implemented via the machine, detection data that includes first portions of the content data associated with a protected attribute and second portions of the content data associated with a predetermined metric; generating, via the machine, analysis data associated with the protected attribute and the predetermined metric; and facilitating display, via the content analysis control panel, the analysis data associated with the protected attribute and the predetermined metric.
Claims20
A method comprising:
- generating, via a machine that includes at least one processor and a non-transitory machine-readable storage medium and utilizing a graphical user interface, a content analysis control panel;
- receiving, via the machine, content data generated via a first at least one AI model;
- generating, via the machine, a virtue scoring model based on training a plurality of second AI models of the virtue scoring model based on training data, wherein the training data is different from the content data, and wherein the plurality of second AI models is different from the first at least one AI model;
- detecting, via one or more third AI models implemented via the machine and different from the first at least one AI model and the plurality of second AI models, detection data that includes first portions of the content data associated with a protected attribute and second portions of the content data associated with a predetermined metric;
- generating, via the machine, analysis data associated with the protected attribute and the predetermined metric;
- generating, via applying the plurality of second AI models of the virtue scoring model, virtue score data for the first at least one AI model based on:
- applying the plurality of second AI models to generate a corresponding plurality of sub-result data based on processing the analysis data;
- and
- generating the virtue score data for the first at least one AI model based on the corresponding plurality of sub-result data generated via applying the plurality of second AI models;
- facilitating display, via the content analysis control panel, of:
- the virtue score data;
- and
- the analysis data associated with the protected attribute and the predetermined metric;
- and
- retraining, based on the virtue score data, the first at least one AI model to generate an updated first at least one AI model.
A system comprises:
- a network interface configured to communicate via a network;
- at least one processor;
- a non-transitory machine-readable storage medium that stores operational instructions that, when executed by the at least one processor, cause the at least one processor to perform operations that include:
- generating, via the at least one processor and utilizing a graphical user interface, a content analysis control panel;
- receiving, via the at least one processor, content data generated via a first at least one AI model;
- generating, via the machine, a virtue scoring model based on training a plurality of second AI models of the virtue scoring model based on training data, wherein the training data is different from the content data, and wherein the plurality of second AI models is different from the first at least one AI model;
- detecting, via one or more third AI models implemented via the machine and different from the first at least one AI model and the plurality of second AI models,
- detection data that includes first portions of the content data associated with a protected attribute and second portions of the content data associated with a predetermined metric;
- generating, via the machine, analysis data associated with the protected attribute and the predetermined metric;
- generating, via applying the plurality of second AI models of the virtue scoring model, virtue score data for the first at least one AI model based on:
- applying the plurality of second AI models to generate a corresponding plurality of sub-result data based on processing the analysis data;
- and
- generating the virtue score data for the first at least one AI model based on the corresponding plurality of sub-result data generated via applying the plurality of second AI models;
- facilitating display, via the content analysis control panel, of:
- the virtue score data;
- and
- the analysis data associated with the protected attribute and the predetermined metric;
- and
- retraining, based on the virtue score data, the first at least one AI model to generate an updated first at least one AI model.
Dependencies are inferred from the claim text.