This workshop explores the societal issues associated with life in a data-driven world where automated forms of information processing affect people's lives in a growing range of ways. Credit card companies, for example, use personal data on purchases for fraud detection. Companies such as Predpol are using past crime data for crime prediction in order to help police departments decide where police should patrol. Schools, employers, and doctors are turning to data-driven systems to shape the future of education, the economy, and healthcare.
That said, it seems likely that our technical capacities have outstripped both contemporary regulatory regimes and public understanding of issues related to privacy, anonymity, bias, and discrimination in data-driven decision making. Looking ahead, technological innovation must be coupled with cultural and historical knowledge to understand the powerful tools we are developing to collect and process information. Liberal arts colleges provide an ideal environment for discussing this complex, interdisciplinary topic.
This workshop will bring together researchers working on the technical and cultural aspects of data analytics to consider key questions related to automated forms of data mining such as: How does access to large data sets and powerful forms of data analytics exacerbate power imbalances? Who is empowered and who disadvantaged? What are the benefits and drawbacks of new forms of social sorting? What forms of regulation might increase accountability and transparency of decision making in the era of big-data?
The hope is that attendees will leave with ideas about how to incorporate material from other disciplines into their teaching and/or their research. This might mean a computer scientist talking more knowledgeably about ethics in a machine learning class, or it might mean a class co-taught by a statistician and a political scientist. It might mean research projects that look at new and emerging forms of data collection, storing, and processing in order to critically evaluate data-driven forms of decision-making.