Systems and methods for automated data extraction of unstructured documents

A method, comprising:

• receiving, by one or more processors coupled with memory, from a client device, an electronic document corresponding to a client account;
• identifying, by the one or more processors, from a plurality of classifications, a classification for at least a portion of the electronic document;
• selecting, by the one or more processors, based on the classification, a set of one or more prompts for one or more artificial intelligence (AI) models according to an extraction plan specifying the set of one or more prompts and a set of one or more mapping rules to extract one or more data elements from the electronic document;
• transforming, by the one or more processors, the one or more data elements extracted based on the set of one or more prompts to map to one or more predefined sets according to the set of one or more mapping rules of the extraction plan, wherein the transformation comprises processing the extracted one or more data elements using a schema that is configured to:
• (i) associate the extracted one or more data elements with corresponding one or more entities, and
• (ii) normalize the extracted one or more data elements according to a canonical format using the set of one or more mapping rules;
• establishing, by the one or more processors one or more relationships between the normalized one or more entities according to the canonical format;
• comparing, by the one or more processors, using the schema and the normalized one or more entities, attributes of the electronic document with attributes of one or more documents corresponding to one or more different client accounts, wherein the one or more documents corresponding to the one or more different client accounts are associated with the classification;
• generating, by the one or more processors, based on the one or more relationships and the comparison of the attributes of the electronic document with the attributes of one or more documents corresponding to one or more different client accounts, a parameter for an extracted data element of the extracted one or more data elements from the electronic document by reference to one or more data elements corresponding to the one or more different client accounts;
• providing, by the one or more processors, for display on the client device, the parameter for the extracted data element corresponding to one of a grade, a score, or a recommendation.

A non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to:

• receive, from a client device, an electronic document corresponding to a client account;
• identify, from a plurality of classifications, a classification for at least a portion of the electronic document;
• select, based on the classification, a set of one or more prompts for one or more artificial intelligence (AI) models according to an extraction plan specifying the set of one or more prompts and a set of one or more mapping rules to extract one or more data elements from the electronic document;
• transform the one or more data elements extracted based on the set of one or more prompts to map to one or more predefined sets according to the set of one or more mapping rules of the extraction plan, wherein the transformation comprises processing the extracted one or more data elements using a schema that is configured to:
• (i) associate the extracted one or more data elements with corresponding one or more entities, and
• (ii) normalize the extracted one or more data elements according to a canonical format using the set of one or more mapping rules;
• establish one or more relationships between the normalized one or more entities according to the canonical format;
• compare, using the schema and the normalized one or more entities, attributes of the electronic document with attributes of one or more documents corresponding to one or more different client accounts, wherein the one or more documents corresponding to the one or more different client accounts are associated with the classification;
• generate, based on the one or more relationships and the comparison of the attributes of the electronic document with the attributes of one or more documents corresponding to one or more different client accounts, a parameter for an extracted data element of the extracted one or more data elements from the electronic document by reference to one or more data elements corresponding to the one or more different client accounts;
• provide, for display on the client device, the parameter for the extracted data element corresponding to one of a grade, a score, or a recommendation.