Monday, April 09, 2007

Lie Detecting Software for Text Communications

The art of money wasting when there's a surplus of research grants and no one to pick them, or a product concept myopia? $680,000 have been awarded by the U.S National Science Foundation to software developers to come up with a lie detecting software for email, IM and SMS messages :

"There's still an open question of whether that is actually possible or not," said Jeff Hancock, a communications professor and information science faculty member at Cornell. "Our research suggests that it is." Passive voice, verb tense changes, and even noun or verb selection can suggest a person is lying, he said. Hancock said another indicator of written deception is the decreased use of the word "I," which is most likely an attempt to create distance. "One of the reasons we think that works as an indicator is that pronoun use is subconscious," he said. In interactive speech, like instant messaging and some dialogues, liars go into a "persuasive mode" and increase the length of their message by 30% to describe and explain situations, he said. Other factors -- such as individual beliefs about behavior, whether someone is accused of something or interacting with an accuser -- can complicate the proces."

Lies are creative even in a written form compared to the favorable body jestures that speak for themselves. And I don't really think an alert such as "the suspect's talking too much on a one sentence question" would do any good. It's all about doing your homework, having experience, not being naive and the power to remain silent when someone's lying to you -- lying pattern intelligence gathering. On the other hand, the product concept myopia is a situation where a company falls in love with their product or service and establish the "build it and they'll come" mentality even without bothering to assess whether or not the market's environment is willing to embrace it, can afford it, or actually need it. The less market transparency, the better for the company, the better the market transprancy the better the puchasing decision of the customer who'll realize that the solution doesn't have to be in the form of the offered product. My point is that, despite the need for the detection of lies of text communications, the solution may not come in the form of talk pattern detection, for instance, your overhyped lover tells you he's in Paris, but geolocating your communicating with him you see he's in Frankfurt, and what a coincidence that is since his ex also lives there.

Using Enron, the infamous case study that'll be discussed in business school for years to come is a good analogy. But just because you think you've established a pattern of communication -- lies -- in conversations that are fake by default, doesn't mean you'll be able to build the dynamics of lying into a detectable pattern. Detecting lies on the fly remains futile for the time being, and you really don't need a program to tell you if someone's lying to you especially in a written form. Outsmart them, act like you don't know to get intelligence on their lying pattern, remain silent for a short timeframe, they'll lie again, be prepared and hopefully you'll recognize a new pattern. Enron's past communication shouldn't be the benchmark in this case, try some Fool's day press releases like this PirateBay announcement for finding a permanent hosting solution - in North Korea! Average people's patterns are the same, therefore pretend to be a moron when you're most knowledgeable, and pretend to be weak when you're most strong and I guarantee you a quick reboot of your relationships.

The lines between sarcasm and a lie are getting even more blurred these days.