Predicting targeted, agency-specific recovery events using trained artificial intelligence processes

A computer-implemented method, comprising:

• receiving, using at least one processor, an identifier associated with an occurrence of a first event from a computing system, and based on the received identifier, obtaining, using the at least one processor, elements of interaction data associated with the occurrence of the first event from a data repository;
• generating, using the at least one processor, an input dataset based on the elements of interaction data;
• establishing, using the at least one processor, a plurality of decision trees for a trained, gradient-boosted, decision-tree process based on corresponding elements of process data, and provisioning, using the at least one processor, a corresponding element of the input dataset to an input node of each of the established decision trees;
• based on the provisioning of the elements of the input dataset to the input nodes of the established decision trees, performing operations, in parallel across a plurality of distributed computing components interconnected across a computing network, using the at least one processor, that apply the trained gradient-boosted decision tree process to the input dataset and that generate, for each of a plurality of candidate event assignments, elements of output data indicative of an expected occurrence of a second event associated with a corresponding one of a plurality of targeted classes of events during a future temporal interval;
• and
• transmitting, using the at least one processor, at least a portion of the generated output data to the computing system, the computing system being configured to perform operations that assign the first event to a corresponding one of the candidate event assignments based on the elements of output data.

A tangible, non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform a method, comprising:

• receiving an identifier associated with an occurrence of a first event from a computing system, and based on the received identifier, obtaining elements of interaction data associated with the occurrence of the first event from a data repository;
• generating an input dataset based on the elements of interaction data;
• establishing a plurality of decision trees for a trained, gradient-boosted, decision-tree process based on corresponding elements of process data, and provisioning a corresponding element of the input dataset to an input node of each of the established decision trees;
• based on the provisioning of the elements of the input dataset to the input nodes of the established decision trees, perform operations, in parallel across a plurality of distributed computing components interconnected across a computing network, that apply the trained gradient-boosted decision tree process to the input dataset and that generate, for each of a plurality of candidate event assignments, elements of output data indicative of an expected occurrence of a second event associated with a corresponding one of a plurality of targeted classes of events during a future temporal interval;
• and
• transmitting at least a portion of the generated output data to the computing system, the computing system being configured to perform operations that assign the first event to a corresponding one of the candidate event assignments based on the elements of output data.