Features

The previous KDnuggets Poll asked: Will software be able to reliably detect lies in a written text in the next 5 years? Based on 280 votes, the results are:

11% -- yes, most of the time
33% -- sometimes, but it will be better than guessing
56% -- No

However Normand Peladeau commented that some software has already been developed that can detecting lies with text mining. Work has been done on this topic since more than 30 years and more recently, James W. Pennebaker from the University of Texas use his content analysis software (LIWC) to detect deception. See the last two papers on page

www.simstat.com/LIWC.htm

In "Lying Words: Predicting Deception from Linguistic Styles," by M. Newman, James W. Pennebaker, Diane S. Berry, and Jane M. Richards write:

In an analysis of five independent samples, a computer-based text analysis program correctly classified liars and truth-tellers at a rate of 67% when the topic was constant and a rate of 61% overall. Compared to truth-tellers, liars used fewer self-references, other-references, and exclusive words and more "negative emotion" and "motion" words.

Here is full paper (in MS word).