Editable rules-based model system

A computerized method for creating and managing an editable rules-based model, the computerized method comprising:

• training, by a processor, a model to generate an output value based on input feature data of a data source;
• generating, by the processor, using the trained model, a feature-output value pair and an associated confidence value for each data source of a plurality of data sources;
• aggregating, by the processor, the confidence values associated with the feature-output value pair for each data source of the plurality of data sources to form an aggregated confidence value of the feature-output value pair;
• building, by the processor, a rule of a rules-based model using the feature-output value pair and the aggregated confidence value;
• providing, by the processor, access to the rules of the rules-based model to a user via a model management interface;
• updating, by the processor, the rule of the rules-based model based on rule change instructions received from the user via the model management interface;
• and
• generating, by the processor, an output value from the updated rule of the rules-based model based on input feature data of a data source.

One or more computer storage media having computer-executable instructions for creating and managing an editable rules-based model that, upon execution by a processor, cause the processor to at least:

• train a model to generate an output value based on input feature data of a data source;
• generate, using the trained model, a feature-output value pair and an associated confidence value for each data source of a plurality of data sources;
• aggregate the confidence values associated with the feature-output value pair for each data source of the plurality of data sources to form an aggregated confidence value of the feature-output value pair;
• build a rule of a rules-based model using the feature-output value pair and the aggregated confidence value;
• provide access to the rule of the rules-based model to a user via a model management interface;
• update the rule of the rules-based model based on rule change instructions received from the user via the model management interface;
• and
• generate an output value from the updated rule of the rules-based model based on input feature data of a data source.