The Linguistics of Status, Influence, and Innovation: A Computational Perspective

Sunday, 15 February 2015: 3:00 PM-4:30 PM
Room LL20C (San Jose Convention Center)
The increasing availability of large-scale text datasets from social media, digitized humanities archives, open government, open-access scientific publications, and crowdsourcing opens the door to an intriguing new possibility: investigation and measurement of language's social function through large-scale computational analysis. By linking socially relevant metadata and annotations with a broad array of linguistic features, ranging from orthography and phonology to discourse structure, researchers can now apply powerful techniques from statistical machine learning to yield new insights on the way language shapes -- and is shaped by -- social interaction. The symposium will feature research spanning a broad range of social phenomena: from macro-scale city-to-city transmission of linguistic innovations, to the linguistic realization of power dynamics in oral arguments before the U.S. Supreme Court, to the performance of interpersonal stances such as friendliness and flirtatiousness in dyadic conversations. In each case, research is shaped and motivated by theoretical models from social science, but leverages new kinds of data and methods that link these theoretical models to semantically meaningful linguistic content. This research also adds a new social dimension to computational linguistics, which has hitherto focused primarily on language-internal structure and on the communication of propositional information.
Jacob Eisenstein, Georgia Institute of Technology
Jacob Eisenstein, Georgia Institute of Technology
Jacob Eisenstein, Georgia Institute of Technology
Sociolinguistic Variation in Online Social Media
Lillian Lee, Cornell University
Language, People, Influence
See more of: Anthropology, Culture, and Language
See more of: Symposia