Text-Based Machine Learning, Big Data, and the Social Study of Science - II

Abstract

Social scientists studying science and technology are eager to exploit the availability of large-scale metadata of scientific publications (i.e., citation networks, authorships, etc.) and corpora (i.e., text as data). The correlate of interest in big data is the development of machine learning techniques. Among the more popular of these techniques is topic modeling, which was originally developed in computational linguistics to explore intellectual currents and structures latent within large-scale text corpora from disparate fields over time and place. Computational linguistics has been variously employed in reconstructing the history of a field (Anderson, McFarland & Jurafsky, 2012; Hall, Jurafsky & Manning, 2008); explaining scientists’ choice of research strategy (Foster, Rzhetsky & Evans, 2015); and modeling scientific discovery (Shi, Foster & Evans, 2015; Rzhetsky, Foster, Foster & Evans, 2015). These approaches remain grounded in steadfast social theories of science and scientific practice. As such, we have the opportunity to corroborate and to complement common approaches in the social study of science, like ethnographies and archival studies. This panel calls for papers that use text-based machine learning techniques to explore the dynamics of knowledge generation, integration, and diffusion in science or technology. It also welcomes related approaches, such as social network analysis and machine learning, or studies that leverage big data in the study of knowledge, science, or technology. Here, we qualify big data as satisfying one of the 3Vs criteria: Volume (e.g., size), Variety (e.g., combined from multiple, disparate sources), and Velocity (e.g., highly granular temporal data).

Chair

• Sebastian Munoz-Najar Galvez, Stanford University

Discussant

• Sebastian Munoz-Najar Galvez, Stanford University

Individual Presentations

• A Critical Examination of Computational Social Science - Karen Y Huang, Harvard University; Peaks Krafft

• Data-Driven Philosophy of Scientific Change - Muhammad Ameer ul Azeem Sarwar, University of Toronto; Patrick Fraser, University of Toronto

• Standing on the Shoulder of the Tumbling Giants: Empirical Analysis on Post-Retraction Citation - seokkyun woo, Georgia Tech

• Making Attention Regimes Visible: An Exploration of Triggers and Patterns of Collective Attention Using Wikipedia - Nicolas Baya-Laffite, STSLab, Université de Lausanne; Boris Beaude, STSLab, Université de Lausanne; Ogier Maitre, STSLab, Université de Lausanne

• Memecracy: Shaping Public Opinion through WhatsApp Data Infrastructure - Henry Chavez, Pontificia Universidad Católica del Ecuador; Jacqueline Gaybor, Erasmus University Rotterdam; Denis Ismael Chavez Ordoñez, PUCE

Session Organizer

• Charles Jonathan Gomez, City University of New York, Queens College