LINGUIST List 23.306

Tue Jan 17 2012

Calls: Computational Ling, Forensic Ling/France

Editor for this issue: Alison Zaharee <>

Date: 16-Jan-2012
From: Eileen Fitzpatrick <>
Subject: EACL 2012 Workshop on Computational Approaches to Deception Detection
E-mail this message to a friend

Full Title: EACL 2012 Workshop on Computational Approaches to Deception Detection
Date: 23-Apr-2012 - 23-Apr-2012 Location: Avignon, France Contact Person: Eileen Fitzpatrick
Meeting Email: < click here to access email >
Web Site:
Linguistic Field(s): Computational Linguistics; Forensic Linguistics

Call Deadline: 03-Feb-2012

Meeting Description:

The EACL 2012 One-Day Workshop on Computational Approaches to Deception Detection explores empirical approaches to detecting deception through linguistic behavior and related modalities. The workshop is part of the EACL 2012 conference to be held in Avignon, France April 23-27, 2012.

The ability to detect deceptive statements has broad applications in law enforcement, business, national security, human resources, advertising, and in predatory communications, including Internet scams, identity theft, and fraud. Deceptive communications may come from a variety of spoken and written sources, including police interviews, legal depositions and testimony, online postings, email, witness and suspect statements, and coded conversations.

The empirical study of deception in language dates at least from Undeutsch (1954, 1989), who hypothesized that 'there are certain relatively exact, definable, descriptive criteria that form a key tool for the determination of the truthfulness of statements'. Reviews from the field of psychology indicate that many types of deception can be identified because the liar's behavior - verbal, visual, and physiological - varies considerably from that of the truth teller's. Even so, humans are notoriously poor at spotting deception, with accuracy rates at the level of chance. Can machines do better?

Several areas of natural language processing are ripe to address the descriptive criteria associated with deception, including text classification, spoken language processing, sentiment analysis, discourse, and pragmatics. New approaches might combine information from different modalities, for example, computational approaches to the analysis of facial expressions may also impinge on the identification of deceptive language. A spate of recent NLP papers on the classification of narratives as true/false suggests that the field is ready to open up to this promising application.

2nd Call for Papers:

The submission deadline has been extended to February 3, 2012.

The workshop on Computational Approaches to Deception Detection, sponsored by the European chapter of the Association for Computational Linguistics (EACL), invites contributions from the NLP community as well as participation from researchers who deal with deception detection from different perspectives, including psychology, neuroscience, and human-computer interaction.


- Classification techniques for identifying deceptive language- Corpora for testing judgments of deceptive language- Corpus annotation for deception cues- Corpus annotation for ground truth- Gathering data from forensic contexts- Online deception- Trustworthiness- Relationships between deceptive language, autonomic responses, and facial expressions- Relationships between deceptive language and neuroimaging- Comparing human to machine performance in deception detection- Portability of deception models to languages other than English- Applications of deception detection- Fraud detection

Important Dates:

February 3, 2012: Paper due dateFebruary 24, 2012: Notification of acceptanceMarch 9, 2012: Camera-ready deadlineApril 23, 2012: Workshop on Computational Approaches to Deception Detection

Page Updated: 17-Jan-2012