Targeted data retrieval and decision-tree-guided data evaluation

An apparatus for data evaluation in accordance with a shared decision tree data object, the apparatus comprising at least one processor and at least one memory including program code, the at least one memory and the program code configured to, with the processor, cause the apparatus to at least:

• generate, using a plurality of feature extraction threads, shared evidentiary data, wherein generating the evidentiary data comprises:
• identify one or more structured feature extraction threads of the plurality of feature extraction threads and one or more unstructured feature extraction threads of the plurality of feature extraction threads;
• cause each structured feature extraction thread of the one or more structured feature extraction threads to extract corresponding per-source retrieved feature data from a corresponding structured data source of one or more structured data sources that is associated with the structured feature extraction thread;
• cause each unstructured feature extraction thread of the one or more unstructured feature extraction threads to (i) extract corresponding per-source unstructured feature data from a corresponding unstructured data source of one or more unstructured data sources that is associated with the unstructured feature extraction thread, and (ii) process the corresponding per-source unstructured feature data for the unstructured feature extraction thread to generate per-source engineered feature data for the unstructured feature extraction thread;
• and
• aggregate each per-source retrieved feature data extracted by a structured feature extraction thread of the one or more structured feature extraction threads and each engineered feature data generated by an unstructured feature extraction thread of the one or more unstructured feature extraction threads to generate the evidentiary data;
• generate, based on a selected evidentiary data subset of the shared evidentiary data that correspond to one or more selected nodes of the shared decision tree data object, refined evidentiary data;
• process the refined evidentiary data in accordance with the shared decision tree data object to generate an evaluation output and an explanation output;
• and
• display an evaluation output user interface comprising user interface data describing the evaluation output and the explanation output.

A computer program product for data evaluation in accordance with a shared decision tree data object, the computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions configured to:

• generate, using a plurality of feature extraction threads, shared evidentiary data, wherein generating the evidentiary data comprises:
• identify one or more structured feature extraction threads of the plurality of feature extraction threads and one or more unstructured feature extraction threads of the plurality of feature extraction threads;
• cause each structured feature extraction thread of the one or more structured feature extraction threads to extract corresponding per-source retrieved feature data from a corresponding structured data source of one or more structured data sources that is associated with the structured feature extraction thread;
• cause each unstructured feature extraction thread of the one or more unstructured feature extraction threads to (i) extract corresponding per-source unstructured feature data from a corresponding unstructured data source of one or more unstructured data sources that is associated with the unstructured feature extraction thread, and (ii) process the corresponding per-source unstructured feature data for the unstructured feature extraction thread to generate per-source engineered feature data for the unstructured feature extraction thread;
• and
• aggregate each per-source retrieved feature data extracted by a structured feature extraction thread of the one or more structured feature extraction threads and each engineered feature data generated by an unstructured feature extraction thread of the one or more unstructured feature extraction threads to generate the evidentiary data;
• generate, based on a selected evidentiary data subset of the shared evidentiary data that correspond to one or more selected nodes of the shared decision tree data object, refined evidentiary data;
• process the refined evidentiary data in accordance with the shared decision tree data object to generate an evaluation output and an explanation output;
• and
• display an evaluation output user interface comprising user interface data describing the evaluation output and the explanation output.