Moya Anegón, F. de; López-Huertas, M.J.: ¬An automatic model for updating the conceptual structure of a scientific discipline (2000)
0.00
0.0015127839 = product of:
0.0030255679 = sum of:
0.0030255679 = product of:
0.0060511357 = sum of:
0.0060511357 = weight(_text_:a in 126) [ClassicSimilarity], result of:
0.0060511357 = score(doc=126,freq=10.0), product of:
0.053105544 = queryWeight, product of:
1.153047 = idf(docFreq=37942, maxDocs=44218)
0.046056706 = queryNorm
0.11394546 = fieldWeight in 126, product of:
3.1622777 = tf(freq=10.0), with freq of:
10.0 = termFreq=10.0
1.153047 = idf(docFreq=37942, maxDocs=44218)
0.03125 = fieldNorm(doc=126)
0.5 = coord(1/2)
0.5 = coord(1/2)
- Abstract
- Knowledge changes and evolves with time. This makes it necessary for classifications and other information retrieval tools to be revised periodically to ensure that they represent an updated organization of knowledge. This problem worsens in the case of some scientific disciplines because of their degree of internal specialization and dynamism. This is the case with Biotechnology and Applied Microbiology (category taken from the Journal Citation Reports (JCR)). A way of solving this problem can be by means of an automatic classification, but also by applying multivariate techniques of analysis and neural networks. This work deals with the two last procedures. The source chosen for doing this work is the Science Citation Index (SCI-CD) database, chosen because it contains the best publications on the disciplines studied - a guarantee for them to be represented properly. We are especially concerned with the quality of the representation. All of the journals dealing with Biotechnology and Applied Microbiology in the database (152) have been selected. Each document has been represented according to the references that it includes. From this point three co-citation matrices were generated: authors, journals and articles from each document. This technique implies the use of complementary methods for document representation by the way of the three matrices processed by multivariate techniques (MDS, PCA, CA). The result is that it can be determined which classes form the intellectual structure of the disciplines studied. This would be of great help for automatic updating of bibliographic classifications and other information retrieval tools. At the same time, these techniques can work as robots for document classification, authors and publications, and allows for the generation of knowledge maps that can be used as interfaces for accessing the documents from their location in a database..
- Type
- a
