performing an extracted text element to a Document Object Model (DOM) element matching operation to identify individual text and DOM elements corresponding to a Portable Document Format (PDF) page of a document which match, said matching operation matching a first extracted text element to a first DOM element;

generating a first synthesized text element including information from said first extracted text element and said first DOM element;

storing said first synthesized text element in a data structure based on information indicating the position at which text included in the first synthesized text element is to be displayed in a rendered image of said PDF page, storing said first synthesized text element in a data structure including storing said first synthesized text element as a node in a k-dimensional (K-D) tree;

receiving information indicating a selection start point in a rendered image of the PDF page and a selecting stopping point in the rendered image of the PDF page; and

using the information in the first synthesized text element to determine which text was selected on said PDF page by a user selection operation, said step of using the information in the first synthesized text element to determine which text was selected on said PDF page including accessing the K-D tree to identify synthetic text elements having rendered image locations which fall fully or partially within a rectangular selection bounding box defined by said selection starting point and said selection stopping point.