Abstract
Social scientists are now in an era of data abundance, and machine learning tools are increasingly used to extract meaning from data sets both massive and small. We explain how the inclusion of machine learning in the social sciences requires us to rethink not only applications of machine learning methods but also best practices in the social sciences. In contrast to the traditional tasks for machine learning in computer science and statistics, when machine learning is applied to social scientific data, it is used to discover new concepts, measure the prevalence of those concepts, assess causal effects, and make predictions. The abundance of data and resources facilitates the move away from a deductive social science to a more sequential, interactive, and ultimately inductive approach to inference. We explain how an agnostic approach to machine learning methods focused on the social science tasks facilitates progress across a wide range of questions.
Highlights
For much of its history, empirical work in the social sciences has been defined by scarcity
This article provides an overview of how social scientists have used machine learning methods, how they have evaluated the performance of models, and what is distinctive about a social science approach to machine learning
We describe our approach to machine learning as agnostic because we avoid assuming that the data emerge from a process that matches our machine learning method
Summary
For much of its history, empirical work in the social sciences has been defined by scarcity. Computation was an even more pressing bottleneck with limited and expensive computing time The consequence of this scarcity was that social scientists developed and relied on statistical techniques that enabled progress with few data and even less computing power. The results include more accurate spam filters and algorithms that can generate realistic fake images, write near-human-quality prose, and defeat world champion human players in games of strategy Just as they have transformed so many other areas of life, machine learning methods have transformative potential in social science. This article provides an overview of how social scientists have used machine learning methods, how they have evaluated the performance of models, and what is distinctive about a social science approach to machine learning. Most of our perspective on the core tasks extends beyond machine learning and draws on the long history of techniques
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
