Systems and methods for generating document templates from a mixed set of document types

A computer-implemented method of generating a template, the method implemented by a template generation server having a memory and a processor, the method comprising:

• receiving a batch of documents including a plurality of documents of different document types;
• identifying a plurality of text blocks located within each document of the batch of documents, wherein each text block includes a text value and a spatial location of the text block within the document, wherein the spatial location is defined by a field box including at least two bounding boxes;
• generating a plurality of clusters, wherein each of the plurality of clusters includes the plurality of text blocks identified in each respective document, wherein each of the plurality of text blocks included in each cluster has substantially matching text values and spatial locations;
• generating a plurality of document arrays, wherein each of the plurality of document arrays corresponds to one of the plurality of clusters and includes a listing of documents containing the plurality of text blocks included in the cluster;
• comparing each document array to each other document array to determine a percentage match;
• when the percentage match between two or more document arrays exceeds a first threshold, defining a preliminary subset of documents including the documents included in the matching document arrays;
• comparing the plurality of clusters corresponding to a matching document array to other plurality of clusters in each other matching document array to determine a document match percentage for each of the preliminary subset of documents;
• defining a final subset of documents by removing any of the preliminary subset of documents having the respective document match percentage below a second threshold;
• and
• for each final subset of documents, generating the template.

A template generation system comprising a computing device comprising at least one memory and at least one processor, wherein the template generation system is communicatively coupled to a database, wherein the at least one processor is programmed to:

• store, in the database, a plurality of templates, each template corresponding to a subset of documents;
• store, in the database, a plurality of matched documents, wherein each matched document includes an identifier matching to one of the plurality of templates;
• store, in the database, a plurality of unmatched documents, wherein each of the unmatched documents do not include an identifier matching to one of the plurality of templates;
• receive a batch of new documents including a plurality of different document types;
• identify a plurality of text blocks located within each new document of the batch of new documents, wherein each text block includes a text value and a spatial location of the text block within the new document, wherein the spatial location is defined by a field box including at least two bounding boxes;
• generate a plurality of clusters, wherein each of the plurality of clusters includes the plurality of text blocks identified in each respective new document, wherein each of the plurality of text blocks included in each cluster has substantially matching text values and spatial locations;
• generate a plurality of document arrays, wherein each of the plurality of document arrays corresponds to one of the plurality of clusters and includes a listing of new documents containing the plurality of text blocks included in the cluster;
• compare each document array to each other document array to determine a percentage match;
• when the percentage match between two or more document arrays exceeds a threshold, define a preliminary subset of documents including the documents included in the matching document arrays;
• compare the plurality of clusters corresponding to a matching document array to other plurality of clusters in each other matching document array to determine a document match percentage for each of the preliminary subset of documents;
• define a final subset of documents by removing any of the preliminary subset of documents having the respective document match percentage below a second threshold;
• and
• for each of the final subset of documents matching one of the plurality of templates stored in the database, apply an appropriate template identifier.