Systems and methods for automated document graphing

A method for document vectorization, comprising:

• in an information processing apparatus comprising at least one computer processor:
• receiving, from a document repository, a plurality of source documents;
• extracting text from each of the plurality of source documents;
• cleaning the extracted text by removing unnecessary characters from the extracted text;
• separating the text into a plurality of parts;
• identifying proper noun vectors in the plurality of parts and mapping the proper nouns to vectors in a vector library;
• converting a plurality of words in the extracted text to lemma form;
• rewriting compound term vectors in the extracted text and mapping the compound term vectors to vectors in the vector library;
• mapping each of the remaining words to a vector in the vector library;
• and
• generating a document vector for the vectors in the document.

A method of document querying using a document graph, comprising:

• in an information processing apparatus comprising at least one computer processor:
• retrieving a query comprising a source document, wherein the source document is associated with a source document vector;
• identifying a locality sensitive hashing bucket for the source document;
• calculating a cosine similarity score for the source document vector and a plurality of document vectors for each document associated with the locality sensitive hashing bucket;
• selecting the documents vector having the closest cosine similarity with the source document vector;
• and
• identifying the document associated with the document vector.