Clustering method for analysis of research fields: Examples of composites, nanocomposites and blends

Abstract

Proper planning of fund allocation by R&D departments of research or industrial institutes may be of key importance to increase of their revenue and future success. However, over years, making such decisions has become notoriously difficult, due to the increasing costs of research and shortening lifetime of products and processes. This is complicated even more in the light of growing specialization of research and exponential growth in number of publications. The existing bibliographic approaches used for structuring fields of research have many drawbacks, e.g. too coarse classification due to lack of the required granularity of information or incoherence of the identified thematic clusters. This paper introduces a method for identification of thematic clusters in a given research domain. The method is based on analysis of high co-occurrence frequency of word sets, which are seeds of thematic clusters. The proposed method is universal and can be applied for analysis of any research field, and the publication data extracted can be further analyzed with respect to future development trends.