Extracting data from documents using proximity of labels and data and font attributes
View Patent ↗A method for identifying information in a document may include analyzing the document for a text block containing a structure element, wherein said structure element is a position, font attribute, or text character; applying a rule based analysis on the text block to identify an adjacent label and field containing a value; and identifying said label and said value as a label and value pair in the document.
1. A computer program product for identifying information in a document, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:
scan a document using optical character recognition;
analyze the scanned document for a text block within the scanned document, wherein the text block contains a structure element, wherein said structure element is a position, font attribute, or text character, and wherein a plurality of label and value pairs within the scanned document are identified based on the contents of the text block within the structure element;
apply a rule based analysis on the text block to identify a label and a field containing a value, wherein data in a text block identified as a header, footer, title, heading, or paragraph is not included in any label and value pair;
wherein in response to a first word in the text block being bold and a second word in the text block being non-bold, then the first word in the text block is the label and the second word in the text block is the value;
wherein in response to the first word in the text block being non-bold and the second word in the text block being bold, then the first word in the text block is the label and the second word in the text block is the value;
wherein in response to the first word in the text block containing only lower case characters and the second word in the text block containing only upper case characters, then the first word in the text block is the label and the second word in the text block is the value;
wherein in response to the first word in the text block being followed by a separator character, and there being a second word after the separator character, then the first word in the text block is the label and the second word in the text block is the value;
wherein in response to a first line in the text block containing a first word in bold followed by a separator character, and a second line in the text block containing a second word in non-bold, then the first line in the text block is the label and the second line in the text block is the value; and
identify said label and said value as a label and value pair in the document.