Across many disciplines, we analyze networks that have been reconstructed or inferred from time series data (e.g., changes in brain activity in neuroscience, shifting stock prices in economics, population dynamics in ecology). These networks can be reconstructed using a variety of techniques, but because different algorithms can output different networks, practitioners are often uncertain about whether their approach is suitable for describing the system in question. In January, NetSI hosted its first annual Collabathon where dozens of researchers worked for three days to start a collaboration and develop a software package, which has since grown to include 19 methods for reconstructing networks from time series data and 21 different network similarity / graph distance measures. Using this software, we compare different techniques for reconstructing networks from time series data. Instead of ranking these methods by their effectiveness at reconstructing networks, we cluster reconstruction techniques based on the similarity of the networks they output (using all 21 similarity / distance measures). The ultimate goal is to provide a map of the various tools that network scientists have at their disposal for studying networks generated from temporal data. This talk will be an update about the current state of the project, a description of the main and secondary results, and an opportunity for members of the NetSI community to weigh in with suggestions, comments, and critiques that will guide the closing stage of this project and improve the final output.
more upcoming events
No upcoming events