An article in PLoS ONE presents a novel methodology for tracking change in large scale networks. Using the bootstrap statistical method, the authors analyze the citation patterns of 7000 scientific journals over the past decade, finding that neuroscience has transformed into a mature and independent field. The authors point out that the analysis of large-scale network activity is integral to many scientific disciplines, from biological to economical. However, despite the great strides that have been made in tracking that activity, there are no significantly powerful tools for mapping changes to the structure of the network itself. Such a tool is critical if we are to continue to ask questions involving network activity; the authors seek to provide such a tool.
The basis of their method relies upon the bootstrap, "a statistic method for assessing the accuracy of an estimate by resampling from the empirical distribution." The power of the bootstrap method lies in its ability to handle data sets whose underlying distribution is not accessible. The authors developed a 4 set process for analyzing the change in a large scale network:
1. Cluster the original networks observed at each time point. 2. Generate and cluster the bootstrap replicate networks for each time point. 3. Determine significance of the clustering for at each time point. 4. Generate an alluvial diagram to illustrate changes between time points.
To demonstrate their method, they apply it to ~ 35 million citations from ~7000 scientific journals over the last 10 years. By using the citation patterns, they seek to track the flow of scientific ideas over time. The alluvial diagram they generate shows striking changes in citation behavior for distinct scientific disciplines, particularly for neuroscience.
In their citation behavior, neuroscientists have finally cleaved from their traditional disciplines and united to form what is now the fifth largest field in the sciences (after molecular and cell biology, physics, chemistry, and medicine). Although this interdisciplinary integration has been ongoing since the 1950s , only in the last decade has this change come to dominate the citation structure of the field and overwhelm the intellectual ties along traditional departmental lines.
The figure above shows an alluvial diagram for a set of biomedical fields for the years 2001, 2003, 2005, and 2007. As someone just starting her career, the obvious, rapid cohesion of citations into a Neuroscience field is thrilling.
The full paper makes for an interesting read, particularly if you are interested in methods of large data analysis, or if you enjoy admiring complex graphical depictions of large data sets.