Systematization of obtaining thematic indicators of scientific information

ABSTRACT

Objective:

In the context of the development of information metric studies, this article proposes and applies a method for obtaining thematic indicators on descriptors representative of themes, subjects or keywords addressed in bibliographic records in the Information Science area.

Method:

A methodological research of applied nature was carried out, using technical procedures of automatic indexing and information metric studies. Initially, a corpus was delimited contemplating a set of bibliographic records referring to 60 articles of Brazilian journals. Later, the software Maui was applied as an automatic indexing system for categorizing the keywords of the bibliographic records into concepts of a specialty thesaurus, contemplating descriptors in the language of the metadata text. Next, the Iramuteq software was applied to generate the thematic indicators from the descriptors obtained by the automatic indexing. Finally, the proposed method was validated based on the analysis of the results obtained for the corpus.

Result:

The flowcharts for validation of automatic indexing and for validation of the information metric study are described, aiming to explain the processes of the proposed method as steps in processing the corpus. In addition, quality metrics for automatic indexing are presented, as well as statistical analyses of word frequency and co-occurrence, and term frequency, where the corpus' main thematic indicators are pointed out.

Conclusions:

We conclude that the thematic indicators obtained by applying the proposed method represent the main themes identified in the corpus, and that the method can be applied to obtain thematic indicators for other sets of bibliographic records.

KEYWORDS:
Information metric studies; Automatic indexing; Thematic indicators; Bibliometrics; Methodological pathway