Cybernetic Hermeneutics: Integrated Human/Computational Methods for Interpretation and Understanding
Illinois Institute of Technology
Time: 12:00 – 1:30 PM
Location: Carnegie 113
Language is commonly viewed as a straightforward channel for communicating meaning—a producer encodes thoughts in language, transmits a text (i.e., speaks or writes), and a consumer receives them (hears or reads) and decodes the text back into thoughts. Failure of communication due to ambiguity or misinterpretation is viewed as an aberration, due to bugs in either language or human information processing. Nothing could be further from the truth—the multiplicity of meanings that language affords is a feature, and the difficulty of understanding language is inherent in the nature of the task. Interpretation is not simply a matter of translating an utterance in a natural language into some “language of thought”, but a more involved process of constructing meanings from the utterance, constrained by its form, but also by the previous knowledge of the consumer (including about the producer), and the broader context of the utterance.
In this talk, I will discuss forays into a “cybernetic hermeneutics”, wherein we seek to devise methodologies enabling productive collaboration between computational systems and human interpreters to develop useful (and empirically supported) understandings of texts, looking specifically at meaning beyond the denotational (the traditional who, what, where, when, and how). This involves devising appropriate computational analytical systems but also methodologies for people to work with these systems effectively. I will describe several case studies, including analysis of scientific methods from the style of the scientific literature, understanding the relationships of different language varieties (registers), forensic authorship attribution, and literary interpretation, and draw out the common themes that point the way forward.
About Professor Shlomo Argamon:
Shlomo Argamon is Professor and Chair of the Computer Science Department and founder of the Master of Data Science Program at Illinois Institute of Technology (Chicago, IL). He received his B.Sc. in Applied Mathematics from Carnegie-Mellon (1988) and his Ph.D. in Computer Science from Yale University (1994), where he was a Fannie and John Hertz Foundation Fellow. He was a Fulbright Postdoctoral Fellow at Bar-Ilan University (1994-96), the Aston University Distinguished Lecturer in Forensic Linguistics in 2014, and is a Fellow of the British Computer Society.
Prof. Argamon’s research focuses on developing computational methods for style-based analysis of natural language using machine learning and shallow lexical semantic representations, exploring applications in intelligence analysis, forensic linguistics, biomedical informatics, and humanities scholarship. His current research work includes developing empirically founded methods for literary analysis of large bodies of text, authorship profiling of anonymous documents, extracting structured representations of opinions from text, and evidence-based analysis of scientific literature. He is particularly interested in elucidating the relationships among linguistic structures, individual goals and reasoning, and social context.
Prof. Argamon has published over 100 scientific articles on machine learning and computational linguistics, is the founding co-chief editor of Frontiers in AI: Language and Computation, and is the co-editor of the books Computational Methods in Counterterrorism (Springer, 2009) and The Structure of Style (Springer, 2010).