System and method for artificial intelligence-based data structuring for document data objects

A system comprising:

• at least one processor in communication with at least one non-transitory computer-readable medium having software instructions stored thereon, wherein the at least one processor, upon execution of the software instructions, is configured to:
• obtain a plurality of documents, each document representing unstructured data;
• for each document, generate a nested database object of a plurality of nested database objects by:
• performing optical character recognition on each document to produce document data associated with each document;
• the nested database object comprising nested database object sections comprising:
• a JobSkillsRequirement section comprising:
• a SkillSet subsection that lists individual required skills, and
• a skillsGroup subsection that encodes logic gating semantics for skill groups;
• and
• generate, for each JobSkillsRequirement section, a concatenated representation of the SkillSet and skillsGroup subsections that preserves the logic gating semantics;
• construct a document similarity graph by computing similarity scores between the concatenated representations and one or more other nested database object sections, wherein each node in the document similarity graph represents a document of the plurality of documents, the document similarity graph being configured to enable scalable management of a plurality of documents in at least one database, and wherein edges between nodes of the document similarity graph are formed where a similarity score between the concatenated representations of the nodes exceed a predetermined threshold;
• utilize at least one graph partitioning model to partition the document similarity graph into clusters to group the plurality of nested database objects into a plurality of document clusters based at least in part on at least one similarity measure of the document similarity graph;
• utilize at least one generative model to generate, for each cluster, from the at least one nested database object, at least one vector embedding representing at least one template document requirement based at least in part on the nested database object sections and the concatenated representation;
• filter a plurality of data profiles based at least in part on the at least one document template requirement to obtain at least one candidate data profile;
• and
• output, via a display, the at least one candidate data profile.