Analysing full text content by means of a flexible co-citation analysis inspired text mining method - exploring 15 years of JASSS articles
Danilo Saft, Volker Nissen · International Journal of Business Intelligence and Data Mining · 2014
This article presents a comprehensible and powerful method to automatically analyse large sets of full text articles on the level of single terms by creating a term connection graph, akin to a sophisticated tag cloud. As a result, it is possible to quickly receive a visual structuring of a research field, which is particularly helpful for those new to a field to get an overview of the field and for researchers as a whole to relate their own work to that of others. Following the design science research approach, we first explain the desired benefit of such a method that includes a flexible and fast way for researchers to compile complete journals and other scientific sources into quickly interpretable graphs, which have a quantitatively objective foundation. Then, we explain the method and its relation to other approaches such as co-citation analysis and text mining. Next, we present in detail the results of applying this method to 15 volumes/years of Journal of Artificial Societies and Social Simulation (JASSS). The article also links to our source code implementations and provides adjustable step-by-step guides so that others may better benefit from and extend this research.