The aim of this doctoral dissertation is to provide a method to study the development of science and the diffusion of scientific ideas through citation analysis. It is claimed that the main motivation for citing is interestingness of the phenomenon. Documents are cited because the phenomena investigated in the documents are of interest to the citer who wants to do further investigations. In this way we see a citation as the result of an interaction between old ideas and new insights. Citation diffusion as a deterministic process is driven by these interactions. The rate of change of citations is the key by which we monitor the knowledge diffusion process.

We define different times series of citation indicators to describe dynamic aspects of science. We continued our investigations by studying knowledge production through integration and diffusion focusing on the notions of diversity and coherence.

Science is an evolutionary process led by the interaction of different ideas on phenomena under study. Interactions between different ideas related to a phenomenon have different characteristics: these ideas may be coherent, but may also conflict with each other. They may stimulate further developments or may - at least temporarily hinder further progress. We map citations to the investigated phenomena, and establish a function reflecting how citations grow in time. Using spline interpolation we are able to detect changes in the timelines of citation data.

The citation history of Nobelist K.C. Kao's main article containing his highly original idea on fiber optics, is used to illustrate how interactions occur during the citation process. We reveal the different diffusion characteristics of its academic and application phase. In this way our work studies the development of science and links this development to citation analysis.