Monitoring scientific developments from a dynamic perspective: Self-organized structuring to map neural network research
✍ Scribed by Noyons, E. C. M. ;van Raan, A. F. J.
- Publisher
- John Wiley and Sons
- Year
- 1998
- Tongue
- English
- Weight
- 291 KB
- Volume
- 49
- Category
- Article
- ISSN
- 0002-8231
No coin nor oath required. For personal study only.
✦ Synopsis
With the help of bibliometric mapping techniques, we tion level choice has been made, the problem of selecting have developed a methodology of ''self-organized'' relevant data within the chosen source(s) arises. 1 Articles structuring of scientific fields. This methodology is ap-(or patents, or documents in general) representing a sciplied to the field of neural network research. We propose ence (or technology) field, are usually selected on the a field-definition based on the present situation. This is basis of key-terms, classification codes, journal names, done by letting the data themselves generate a structure, and, with that, define the subdivision of the research field authors names, or author affiliation addresses. Often, an into meaningful subfields. In order to study the evolution iterative process is applied: Documents selected, for inover time, the above ''self-organized'' definition of the stance, by key-terms yield, in turn, other (probably less present structure is taken as a framework for the past central) terms, which are then used to extend the selection structure. We explore this evolution by monitoring the of documents in order to cover the field more widely.
interrelations between subfields and by zooming into the internal structure of each subfield. The overall Then, after the last selection step, the whole set of (''coarse'') structure and the detailed subfield maps documents has to be structured in order to make the data (''fine structure'') are used for monitoring the dynamical accessible and manageable. This structure should be such features of the entire research field. Furthermore, by dethat the component parts provide a meaningful division of termining the positions of the main actors on the map, the field, representing research subfields and application these structures can also be used to assess the activities of these main actors (universities, firms, countries, etc.).
areas. This is particularly important for evaluative pur-Finally, we ''reverse'' our approach by analyzing the deposes, for instance to assess the role and position of actors velopments based on a structure generated in the past.
(countries, universities, companies) in the field (see, for
Comparison of the ''real present'' and the ''present coninstance, Grupp, Schmoch, & Koschatsky, 1997).
structed from the past'' may provide new insight into
If one starts with a document as a unit of information, successful, as well as unsuccessful, patterns and ''trajectories'' of developments. Thus, we explore the poten-there are several ways one can obtain a structure of scitial of our method to put the observed ''actual'' developence. They are all related and are based on specific characments into a possible future perspective.
teristics of the document. These characteristics are, for instance, the journal in which the document is published, the references given in the document, and the document's 1. Introduction: Analysis of the Structure of keywords or classification codes.
Science and Technology
Structures always arise because the composing ele-An important question in the analysis of scientific and ments have particular linkages, indicating degrees of retechnological developments is the following: How can latedness. Here, we have a similar principle: Documents one define and delineate a particular field of science and appear in the same journal, or they have a smaller or technology? Nowadays, there is a large universe of bibliolarger number of references, keywords, or classification graphic databases and other document-related data (van codes in common. Typical bibliometric techniques such Raan, 1996). The Internet makes this universe ever-exas co-citation and co-word analysis are based on this prinpanding. Thus, the first problem is selection: The choice ciple (Callon, Courtial, & Turner, 1983, 1991; Healey, of an appropriate data source. After this higher aggrega-1 An extensive study on this matter is performed by .