SYSTEMS AND METHODS FOR GENERATING ISSUE LIBRARIES WITHIN A DOCUMENT CORPUS

摘要

Method and system of generating an issue library from a document corpus defining a citation network are provided. The method comprises semantically linking documents within the document corpus by: locating at least one citation present in a citing document; determining a reason-for-citing for the at least one citation; determining the cited-text-area present in a cited document, wherein the cited-text-area corresponds with the reason-for-citing; marking boundaries of the reason-for-citing within the citing document; and pairing reasons-for-citing in citing documents with cited-text-areas in cited documents. A cited-text-area in a cited document is a text area that has a highest similarity value of text present within the cited document. A group of semantically-similar reasons-for-citing and cited-text-areas that are semantically similar to at least one issue is created. Information regarding groups of semantically-similar reasons-for-citing and cited-text-areas in an issue library metadata entity, which is associated with an individual issue, is stored.