Recognising document components in XML-based academic articles

Abstract

Recognising textual structures (paragraphs, sections, etc.) provides abstract and more general mechanisms for describing documents independent of the particular semantics of specific markup schemas, tools and presentation stylesheets. In this paper we propose an algorithm that allows us to identify the structural role of each element in a set of homogeneous scientific articles stored as XML files.

Research Articles in Simplified HTML: a Web-first format for HTML-based scholarly articles

“…Also, thanks to the regularity they provide, it is possible to perform easily complex operations on pattern-based documents even when knowing very little about their vocabulary (automatic visualisation of document, inferences on the document structure, etc.). In this way, designers can implement more reliable and efficient tools, can make a hypothesis regarding the meanings of the document fragments, can identify singularities and can study the global properties of a set of documents, as described in Di Iorio et al (2012) and Di Iorio et al (2013).…”

The Document Components Ontology (DoCO)

“…The creation of DoCO was undertaken by studying different corpora of documents (mainly scientific literature and web documents on different topics) and publishers' guidelines, from two perspectives -the structural and the rhetorical -as was also done by past works on document patterns [13][14][15]. We also undertook some informal interviews with researchers in different fields and with academic publishers, in order to gather as much information as possible about document components and their use.…”

“…The regularity of pattern-based documents (defined by means of markup languages such as DocBook or LaTeX) then makes it possible to perform complex operations easily, even when knowing very little about the documents' markup vocabulary. This in turn enables designers to implement more reliable and efficient tools [14], make hypotheses regarding the meanings of document fragments [15], identify special cases, and study global properties of sets of documents [13].…”

“…Note that even if it is not explicitly stated in DoCO, we consider these DC Terms object properties to be transitive. 32 Potentially there exist two different ways of organising footnotes, since their structural semantics can depend on the particular (markup) language we use to express it, as discussed in [15]. The first, is a container-based behaviour, as adopted by JATS [24], that allows one to specify footnotes (through the element ft) by using an element that is totally separated from the main text from which it is referenced (usually through XML attributes).…”

“…In making steps towards addressing this issue, we have recently used DoCO as a theoretical base for the development of an ontology-aware algorithm to retrieve the meaning of markup structures in XML article sources [15], without explicitly looking either at the particular markup language used, or the actual content of the document. The algorithm was developed by starting from the actual specification of DoCO classes, and then tuned according to other statistical and topological principles (e.g.…”