Social scientist (or possibly mad professor) Dan Zarella has now invented what could be a particularly contreversial Twitter application: TweetPsych.
Taken from the main site:
TweetPsych uses two linguistic analysis algorithms (RID and LIWC) to build a psychological profile of a person based on the content of their tweets. The service analyzes your last 1000 tweets and works best on users who have posted more than 1000 updates. It also works best on accounts that are operated by a single user and use Twitter in a conversational manner, rather than simply a content distribution platform.
There’s a bit more detail on his blog, going into what RID and LIWC actually are.
The interesting bit is that compared to most other forms of writing and communication, Twitter is probably the closest to being able to express yourself with little or no thought – it’s simple, quick, and you can hide yourself behind a fake persona or a protected account.
The worrying thing is that it opens up your Twitter account to analysis by friends, families and employees with little or no knowledge of linguistics and profiling.
In fact, the first comments on Dan’s post are discussing using NLP to influence people to puchase via advertising, the difficulty of interpreting the results, and the fact that it could easily be misused.
And without casting any aspersions on Dan’s skills, the possibility of an error appearing in an automated system isn’t unheard of…
Without being able to intepret my results effectively, feel free to explain them to me! Is it good that ‘Positive Emotions’ comes in at No.5, or bad that No.14 appears to be ‘Sad’.
