The recognition of scientific paper titles that don’t correspond to their contents by means of the occurrence of key-words

Authors

  • Manoel Palhares Moreira Departamento de Ciencia da Computação, Pontificia Universidade Católica de Minas Gerais, Brasil
  • Sergio Murilo Stempliuc Departamento de Ciencia da Computação, Pontificia Universidade Católica de Minas Gerais, Brasil

DOI:

https://doi.org/10.54886/scire.v12i1.1597

Abstract

Amethodology for detecting titles that are not reliable surrogates of the content of scientific articles is proposed by comparing them with the keywords or descriptors of the article. It is based on the hypotheses that papers with at least one of their keywords included in the title have a corresponding content. 300 articles from two Brazilian web-magazines were used. First, the occurrence of keywords was checked: in 23% of the cases (69 papers) no keyword was found in the title. Such checking in the other texts showed that 13% (9 texts) had none of the keywords in the text; in 36% (25 texts) they were not found in the abstracts; and, in 58% (40 texts), in the bibliographical references. Next, a qualitative analysis of data was carried out. It was verified that in some cases the problem was keywords indication. As a proposed solution, a checking routine was done, for the frequency of all words used in the papers but stop-words. After that, came the checking of the word variation in the indicated keywords. The conclusion was that keywords in paper titles is a factor that may be observed during the submission process for the checking of title with paper content, and that digital magazines might control vocabulary built form such words. That measure would make it easier for authors in the title creation and keyword indication. Such procedures would facilitate search and finding for documents before users’ needs

Downloads

Download data is not yet available.

Published

2006-06-30

How to Cite

Palhares Moreira, M., & Murilo Stempliuc, S. (2006). The recognition of scientific paper titles that don’t correspond to their contents by means of the occurrence of key-words. Scire: Knowledge Representation and Organization (ISSNe 2340-7042; ISSN 1135-3716), 12(1), 225–245. https://doi.org/10.54886/scire.v12i1.1597

Issue

Section

Articles