Bastián González-Bustamante presents his research at SLAS 2022 conference

We are excited to announce that our research associate, Bastián González-Bustamante, has successfully presented his research paper “Resignation Calls and Ministerial Turnover in Latin America: Notes on Data Gathering using Machine Learning” at the Society for Latin American Studies (SLAS) annual conference held in Bath, UK, on 21-22 April 2022.

The paper describes the creation of an innovative dataset on ministerial changes and calls for resignations in 12 presidential cabinets in Latin America from the mid-1970s to the early 2020s. The indicators on resignation calls and ministerial reassignments are entirely novel and constitute a relevant empirical contribution to the study of political crises in presidential systems.

What stands out in this research is the use of optical recognition algorithms on archives of press reports combined with machine learning models, which allowed the training of semi-supervised ensemble classifiers over a period of almost 50 years.

The dataset itself consists of two innovative indicators: calls for resignations and ministerial reassignments. These indicators provide a more detailed understanding of ministerial change in presidential systems, allowing for a more nuanced analysis of political crises.

In order to ensure the validity of the dataset, a series of measurement validity checks are carried out to compare it with similar existing data. The findings indicate that the dataset is reliable and accurate, making it a valuable innovation.

In addition to presenting the research paper, Bastián González-Bustamante also received recognition for his excellent work in the field of comparative politics and methodology. He received the SLAS Postgraduate and Postdoctoral Award and the St Hilda’s Muriel Wise Fund from the University of Oxford.

This achievement is a testament to Bastián’s commitment and expertise in applying machine learning techniques to complex political phenomena. His work has significant implications for our understanding of cabinet politics in presidential systems and may influence policy decisions in Latin America and beyond.

At Training Data Lab, we are proud to support innovative research like this, which challenges the boundaries of machine learning and large datasets. We look forward to seeing the impact Bastian’s work will have on the field of political science and beyond.

About the research associate

Bastián González-Bustamante is a doctoral researcher at the University of Oxford and a research associate at Training Data Lab, where he works on projects related to machine learning and natural language processing. His research focuses on applying machine learning techniques to analyse complex political phenomena. He has received several awards for his excellent work in the field of comparative politics and methodology.

* AI-generated text
Read more about how we generate our content