RAKE algorithm application in digital document indexing

Currently there are a wide range of digital documents, particularly text documents that by their nature allow automated processing, organization and analysis methods for the purpose of retrieving information concisely and efficiently. Diverse areas of study such as computer science, library science, computational linguistics and text mining, among others, have developed digital document processing methods for the purpose of facilitating their representation, organization and retrieval in digital libraries, databases and catalogs. These methods are both statistical and linguistic in nature. In this paper, the RAKE statistical method is examined in order to identify and extract multiword keywords from digital documents to allow organization, retrieval and automated document indexing.