Given a document, is it an original?
This task is divided into source retrieval and text alignment. Source retrieval is about searching for likely sources of a suspicious document. Text alignment is about matching passages of reused text between a pair of documents.
Given a document, who wrote it?
This task focuses on authorship verification and methods to answer the question whether two given documents have the same author or no. This question accurately emulates the real-world problem that most forensic linguists face every day.
Given a document, what're its author's traits?
This task is concerned with predicting an author's demographics from her writing. Besides being personally identifiable, an author's style may also reveal her age and gender.
Cindy K. Chung
The University of Texas at Austin
How do the words we use reflect our personalities, psychological states, and social relationships? Psychologists have applied simple word counting and word patterning methods such as Linguistic Inquiry and Word Count (LIWC) and the Meaning Extraction Method (MEM) across a variety of contexts to understand individual differences in communication. Word use and interaction patterns in everyday conversations and social media exchanges can reveal personality, engagement, and deception. Quantitative text analysis from a psychological perspective can help to uncover how people connect, communicate, and interact in digital spaces.
Cindy K. Chung is a Postdoctoral Fellow in the lab of Professor James W. Pennebaker in the Department of Psychology at The University of Texas at Austin. She examines how word use reflects personality and social dynamics. Her research involves the development and application of computerized text analysis tools and methods for the social sciences.
Brooklyn Law School
Many forensic sciences have been under attack for their failure to be grounded in adequate research. Even fingerprint evidence is questioned, not for its being generally unreliable, but because there is inadequate research into the extent that accuracy decreases as the quantity and quality of information are each systematically degraded. At the same time, psychological research has revealed biases to which experts are subject, especially confirmation bias – the propensity to value evidence that confirms one’s hypothesis more than evidence that disconfirms it. Somewhat at odds with these observations are interesting debates about the relative accuracy of expert judgment by individuals based on their experience on the one hand, and algorithmic expertise on the other. This presentation will discuss how legal systems react to offers of linguistic expert evidence in light of these developments.
Lawrence M. Solan is the Don Forchelli Professor of Law and Director of the Center for the Study of Law, Language and Cognition at Brooklyn Law School. In October 2012, the Center convened a workshop on authorship attribution, funded by the National Science Foundation, which brought together legal scholars, computer scientists, linguists and statisticians to assess the state of the field in relation to legal standards for scientific evidence. Solan’s writings address such issues as statutory and contractual interpretation, the attribution of responsibility and blame, and the role of the expert in the courts. Solan has been a visiting professor at the Yale Law School, and in the Psychology Department and Linguistics Program at Princeton University.