New method of analyzing networks reveals hidden patterns in data


Scientists have actually developed a new method of determining how relationships in a network modification gradually can expose crucial information about the network. Credit: Alina Grubnyak, Unsplash

A new method of determining how relationships in a network modification gradually can expose crucial information about the network, according to scientists at Penn State and the Korean Rural Economic Institute. For instance, when used to the world economy, the method found the best quantity of network modification throughout 2008-2009, the time of the worldwide monetary crisis.


“Most existing approaches only capture relationship changes in a network one network member at time,” stated Stephan Goetz, teacher of farming and local economics, Penn State, and director of the Northeast Regional Center for Rural Advancement (NERCRD). “Our measure allows us to see how these relationships change over time across the entire network, which will give us potential new insights into how networks behave, as well as the impacts of those changes.”

According to Goetz, networks of all types are represented aesthetically by private nodes linked to one another by lines, or links. The familiar airline company hub-and-spoke map is one example. Each node represents an entity in a network—an airport, an individual, an organisation, or a nation, for instance. The links in between 2 nodes represent their connection or relationship.

“Think about your own network of relationships, with each person in your network represented by a node and your connection to them represented by a link. Over time, some people might drop out, others might come in, some relationships get stronger, others weaken,” Goetz stated. “Because most networks change over time, this configuration of nodes and links also changes.”

This modification from one duration to the next is represented by a modification in the angles formed by the nodes and links, and these angles are the focus of the research study, which was released on July 24 in PLOS ONE.

Goetz and his co-author, Yicheol Han, a research study fellow at the Korea Rural Economic Institute in Nasu-si, South Korea, and previously a research study partner at NERCRD, relied on a familiar mathematical step called the cosine resemblance, which is typically utilized to determine the orientation of the angles in between points and lines radiating from the points. When used to network science, cosine resemblance permitted the scientists to determine the size of the modification in any provided relationship, both relative to itself and to the general network.

To evaluate their new step, the scientists used it to numerous real-world networks, consisting of the World Input-Output (I/O) Table, which records financial deals on an annual basis both throughout and within countries. Concentrating on the years 2000-2014, they discovered that the best quantity of network modification, which they describe as “rewiring,” took place throughout 2008-2009. This was not unforeseen, due to the Global Financial Crisis which took place throughout those years.

“Rewiring is another way of thinking about reorganization, in this case,” Goetz stated. “Our findings show what a tremendous shock the world financial crisis was. It’s interesting that the measure picked this up so strongly, and that it can be used as a new way of quantifying how the economy adjusted after the shock.”

Their step likewise revealed a noticable down pattern in rewiring after 2010, which they assume might have added to the uncommonly sluggish healing from the economic downturn.

Next, the group took a look at 3 private nations—the U.S., Germany and China—to see how the rewiring in these nations’ economies compared. While the U.S. and Germany followed a comparable pattern as the general world economy, China showed a substantially various pattern, with a spike in its rewiring taking place both prior to and after the monetary crisis. Nevertheless, to Goetz’s surprise, they discovered that the monetary crisis was not the most substantial shock to the Chinese economy throughout this time duration.

“The rewiring measure shows that joining the World Trade Organization in 2003 was a bigger shock to the Chinese economy than the Great Recession,” Goetz stated. “I’m not aware that anybody has shown this using any other measure.”

Goetz and Han likewise took a look at how rewiring in numerous countries’ financial I/O tables impacted their substance yearly earnings development rate per capita, the “holy grail” of financial steps, according to Goetz.

“We found that whether countries belonging to the Organization for Economic Cooperation and Development rewired a lot or a little didn’t make much difference to their compound annual growth rate,” stated Han. “On the other hand, former Soviet Union nations really benefitted from rewiring. The more they rewired, the more their annual growth rate increased. We think it’s because they only recently became market economies and lack mature institutions and other adjustment mechanisms to recession, so rewiring has more of an impact on their growth.”

Future research study will take a look at how the step can be used to rural economies, to see how these locations might or might not take advantage of rewiring.

“That has the potential to lead to new ways of looking at how rural areas in the U.S. can develop and adjust to emerging shocks and other structural challenges,” Goetz stated.


Poor psychological health days might cost the economy billions of dollars


More details:
Yicheol Han et al. Determining network rewiring gradually, PLOS ONE (2019). DOI: 10.1371/journal.pone.0220295

Offered by
Pennsylvania State University

Citation:
New method of analyzing networks reveals hidden patterns in data (2019, September 10)
recovered 10 September 2019
from https://phys.org/news/2019-09-method-networks-reveals-hidden-patterns.html

This file goes through copyright. Apart from any reasonable dealing for the function of personal research study or research study, no
part might be replicated without the composed approval. The material is offered details functions just.



Recommended For You

About the Author: livescience

Leave a Reply

Your email address will not be published. Required fields are marked *