Big Data in Politics: Alumnus Eduardo Albrecht
After graduating in International Affairs and History from JCU in 1999, Eduardo Albrecht earned his M.A. in Anthropology of Development and Ph.D. in Anthropology at the School of Oriental and African Studies University of London. He spent six years in Italy working in the nonprofit sector, where he founded and directed ‘Rising South’ until 2008. ‘Rising South’ is a Civil Rights Organization that raised one million Euro in donations annually, and promoted awareness on social/political issues in Europe including migration, environment and civil rights. Eduardo currently teaches in New York City. In recent years Eduardo has been involved in JCU through his affiliation to the Guarini Institute.
A recurring theme in your work is the study of international protest movements. You conducted fieldwork among anti-globalization protesters in Southern Italy and North Korean defectors in South Korea. What’s one finding that surprised you?
First, that revolution is not what it used to be. The level of state control, thanks in part to modern communication technologies, is such that revolution has become, like Guy Debord prophesied in his 1967 book Society of the Spectacle, something of an entertaining pastime, much like politics in general. This has prompted me to study things like social media, big data, machine learning, and artificial intelligence, and how these interact with contemporary both contentious and non-contentious political processes.
You have been experimenting with specifically designed computer programs to correlate patterns in protestors’ online behavior with real-world events. How is technology and artificial intelligence helping your work/research?
Quite a lot. I am currently working on a book project that investigates the possibility of forecasting political crises, such as anti-government protests in unstable nations, by identifying recurring patterns of positive/negative sentiment in a range of different “online milieus” and use these to create leading indicators for political developments in real time. Examples of online milieus include social media websites, instant messaging services, issue-specific blogs, or the comments section of articles in the mainstream media. This mixed methodology project uses natural language processing software to sieve through large amounts of information. Once information is collated, machine learning software is employed to find recurring patterns in past online behavior and real-world political events. Identified patterns are used to generate leading indicators in real time. The software is constantly re-calibrated using ethnographic considerations arising from ongoing fieldwork, consisting of on-site or virtual interviews.
In your opinion, does the use of technology and artificial intelligence have a negative side?
I am going to be a little provocative, and say no. There is much fear out there because A.I. is essentially an unknown entity to most. We conjure up images of an evil, self-aware artificial intelligence system from science fiction movies like The Terminator or The Matrix. And pundits out there like to scare us with the singularity. But the truth is that in my years researching this technology I have found that there is nothing substantially new or particularly dangerous. Certainly there is a lot of data out there on you that you are not aware of, and there are major issues of privacy and control that need to be dealt with, particularly as to what kind of entity (corporate or state) should have this right. But in essence, A.I. is just a very large calculator. The programing behind the algorithm remains human.
You are Founder and Research Director at Ethnographic Edge, a project that tracks developments in international politics. Can you tell us a bit more?
This has recently evolved into a consultancy project named Peloria. We chose the name because it means “an abnormal often hereditary regularity of structure occurring in normally irregular flowers,” and this encapsulates well the idea that you can find patterns where you least expect them. In sum, I have partnered with political economists and data scientists to design software that produces a combined quantitative/qualitative approach that achieves both speed and depth of analysis, where forecasts can be made available to organizations working in unstable geographic areas. Forecasts are expressed as percentages of possibility and are bound to specific time ranges (based on past, recurring patterns). This contributes to the development of timely policy positions and helps organizations minimize potential risks to their operations.
What’s your advice for graduates today?
A graduate degree is a must these days, so I would advise all students to choose their graduate school wisely and to make sure that it aligns with their skills and career goals. I strongly encourage traveling, and living for long periods of time in exotic places, as far away as possible from all their comfort zones! I advise alumni to stay as much in touch with their Alma Mater as possible, particularly when they are in a position to give back. This does not have to be in terms of donations, it can also be in service and participation to the intellectual life of JCU, which I think in recent years has become increasingly rich and interesting. Personally, I am grateful to the Guarini Institute for allowing me to remain intellectually committed and connected with the life of the university.