The nature of an individual document is often defined by its relationship to selected tasks, societal values, and cultural meaning. The identifying features, regardless of whether the document content is textual, aural or visual, are often delineated in terms of descriptions about the document, for example, intended audience, coverage of topics, purpose of creation, structure of presentation as well as relationships to other entities expressed by authorship, ownership, production process, and geographical and temporal markers. To secure a comprehensive view of a document, therefore, we must draw heavily on cognitive and/or computational resources not only to extract and classify information at multiple scales, but also to interlink these across multiple dimensions in parallel. Here we present a preliminary thought experiment for fingerprinting documents using textual documents visualised and analysed at multiple scales and dimensions to explore patterns on which we might capitalise.
展开▼
