Identifying unchecked criteria in unstructured and semi-structured data

A computer-implemented method for identifying unchecked criteria in unstructured data within a form, comprising:

• identifying checked data in the form;
• identifying a first set of unstructured data as pertinent to the checked data in the form;
• performing section identification and annotation operations on the form to identify and categorize sections within a target text of the form;
• and,
• identifying unchecked criteria in unstructured data within the form based upon the identifying the first set of unstructured data, the identifying comprising detecting and classifying text spans representing the unchecked criteria within the form to facilitate accurate interpretation of the text spans, the unchecked criteria comprising a discrete item that represents a question in the form lacking a response;
• training a Question and Answer system based upon the identifying unchecked criteria.