A method and system for computer-processing quotations in digital text to determine an author associated therewith is disclosed. The method comprises receiving a plurality of digital texts. The plurality of digital texts are parsed to extract one or more quotations. At least one candidate authors are identified for each of the digital text. A quotation similarity value for a given quotation with respect to each of the remaining one or more quotations is assigned. A quotation cluster is generated, which comprises one or more similar quotations and a set of candidate authors. The set of candidate authors is analyzed to identify a given candidate author meeting a condition. The candidate author meeting the condition is stored as the author of the one or more similar quotations.