| Abstract | FROM SCIENTIFIC COMMUNICATION TO PUBLIC KNOWLEDGE: THE SCIENTIFIC ARTICLE WEB PUBLISHED AS A KNOWLEDGE BASE
Carlos H. Marcondes
Information Science Department, Information Science Post-graduate Program, Universidade Federal Fluminense
R. Lara Vilela, 126 – 24210-590, Niterói – RJ, Brazil
marcon@vm.uff.br
Paper submitted to ElPub 2005
ABSTRACT
1.Problem. Linking Electronic publishing to Web ontologies should be a
cognitive tool of which its impacts and possibilities are far from
being evaluated. This paper discusses the potential of electronic
publishing of scientific articles not only as text, but also as a
machine readable knowledge base for scientific communication. The
scientific communication process, a slow social one, has been the
traditional mechanism in which new knowledge is incorporated to the
body of Science. The potential of new information technology has been
applied to modern bibliographic information systems to improve
scientific communication, which provides fast notification and
immediate access to full-text scientific documents. Despite these
advances, the scientific communication process depends largely on text
producing, reading and interpretation, and the citation of scientific
articles by researchers. Today, except for some pioneer initiatives,
the majority of the Web-published scientific journals are strongly
based on the paper print publishing model. 2. Objectives. The objective
of this research is to investigate the potential of the Web published
scientific articles, conceived not only as text, but also as a machine
readable knowledge base, explicitly and formally related to a Web-based
public ontology, that represents the assented knowledge of a specific
domain. The goal is to enhance Web electronic publishing to embody new
facilities provided by the Web environment in the Semantic Web
initiative context. There will be presented a framework of a system
that, along with addressing text producing and publishing, will also
capture the structural parts of a scientific article, such as the
statement of the problem, the hypothesis, the methodology and the
conclusions. Then this structure will be represented in a
machine-readable format, and linked to the concepts expressed therein
to a public Web ontology representing the assented knowledge of a
scientific area. 3. Methodology. The article develops a prospective
survey to identify similar proposals and innovative experiences in
electronic publishing of scientific articles, authoring tools to assign
comments to scientific articles, citation analysis and the reasons for
citation. Scientific methodology is also reviewed looking for
structural characteristics of the scientific method and scientific
argumentation. It also reviews different experiences in developing
Markup Language to some specific areas of knowledge, such as Chemical
Markup Language, Mathematics Markup Language, Biology Markup Language.
4. Conclusions. A model of an electronic publishing process is
outlined. This permits the electronic publishing of not only scientific
articles as text, but it also enables an author to formalize the “deep
structure” of a scientific article. This consists of assumptions,
hypothesis, methodology, citations, datasets used, conclusions and
contributions; all of which to publish as a knowledge base, as an
ontology. XML is suggested as the mechanism to represent the structural
characteristics of all these components of a Scientific article, thus
outlining a Sm-ML, a Scientific methodology Markup Language. A Sm-ML
should use the name-space facilities of XML to incorporate different
name-spaces, a generic name-space concerning Scientific methodology and
specific name-spaces for different knowledge areas. Concepts expressed
in the different parts of a Scientific article should also be linked to
public Web ontologies thus enabling the establishment of a formal
relationship between the Scientific article specific ontology to
ontologies like the UMLS – the Unified Medical Language System
(http://www.nlm.nih.gov/pubs/factsheet/umls.html). The citations of an
article should also be linked to the cited Web published scientific
articles as qualified citations, in which the reasons to cite and the
relationship between this specific scientific article and its citations
are made explicit. The proposed model of electronic publishing can
enhance the scientific communication process, permitting semantic
retrieval, critical inquiring, semantic citation, comparison, coherence
verification and validating a scientific article through Web public
ontologies which express the assented knowledge of a scientific area.
The model was also conceived as the base for developing enhanced
authoring and retrieval tools.
Key-words: electronic publishing; scientific communication; scientific
methodology; Semantic Web; XML; medical web ontologies; authoring
tools; e-science
|