New this term, University of Waterloo graduate students will discuss their complex systems work during a monthly seminar. Join us for a 20 minute talk, followed by a 20-minute discussion and feedback opportunity. Afterwards, everyone who is able is welcome to meet at the Graduate House for food and socializing from 6 p.m. onward.
This month, Mahta Ramezanian will discuss 'Political Corruption Networks' on Thursday, June 20th from 4:30 - 5:30 in STC 1019. We hope to see you there!
Mahta Ramezanian is currently a Master's student doing reaserch on Polymer Physics. She holds a bachelor degree is in physics from the Sharif University of Technology in Tehran, Iran, as well as a minor degree in Energy Engineering.
Political Corruption Networks
Ollin D. Langle Chimal & Sonya Ahamed, University of Vermont, Vermont, USA; Mahta Ramezanian, University of Waterloo, Waterloo, Canada, David Sutherland Blair, Universitat Pompeu Fabra, Barcelona, Spain, José R. Nicolás Carlock, National Autonomous Univeristy of Mexico, Mexico City
The rationale for our project is drawn from the United Nations Convention against Corruption (UNCAC- 2003) that acknowledges corruption as a transnational phenomenon that affects all societies in deep and multiple ways, at political, economic, ecological and social fronts. Therefore, comprehensive and multidisciplinary approaches are recommended to prevent and combat the complexity of corruption effectively. For our project we focused on a particular case of corruption in Mexico: the phantom companies of ex-governor of the state of Veracruz, Javier Duarte. This is a recent and paradigmatic case of documented political corruption in Mexico that involves a complex network of hundreds of shell companies used to embezzle billions of dollars. We made use of many qualitative and quantitative methods to study the structure and time-evolution of the system. Networks were built according to the available information and data. The data-set contained information about the relationship among 356 shell companies and 357 people, as well as information about the date of creation of each company. Two types of nodes (companies and people) and five edge categories between those two types of nodes were defined: administrators, legal representatives, share holders, commissioners, and notaries. A bi-partite approach was used to obtain different metrics as function of time and/or edge-type, such as: edge density, clustering coefficients, eigenvalue distributions, assortativity, betweeness and eigenvector centrality, and other ad-hoc quantities. Some commonalities and differences between this case and other cases of criminal networks reported in the literature were checked in order to find a possible ‘signature for corruption networks’.