Mapping change in scientific specialties: a scientometric case study of the development or artificial intelligence

Has an identifiable core of activities called AI been established, during the AI-boom in the eighties? Is AI already in a “paradigmatic” phase? There has been a lot of disagreement among commentators and specialists about the nature of Artificial Intelligence as a specialty. This makes AI an interesting case of an emerging specialty. We use aggregated journal-journal citations for describing Artificial Intelligence as sets of journals; factor analytic techniques are used to analyze the development of AI in terms of (an emerging) stability and coherency of the journal-sets during the period 1982–1992. The analysis teaches us that AI has emerged as a set of journals with the characteristics of a discipline only since 1988. The thereafter relatively stable set of journals includes both fundamental and applied AI-journals, and journals with a focus on expert systems. Additionally, specialties related to artificial intelligence (like pattern analysis, computer science, cognitive psychology) are identified. Neural network research is a part neither of AI nor of its direct citation environment. Information science is related to AI only in the early eighties. The citation environment of AI is more stable than AI itself.