US Secret Service Wants To Identify Snark 213
beschra (1424727) writes "From the article: 'The U.S. Secret Service is seeking software that can identify top influencers and trending sets of social media data, allowing the agency to monitor these streams in real-time — and sift through the sarcasm. "We are not currently aware of any automated technology that could do that (detect sarcasm). No one is considered a leader in that,'" Jamie Martin, a data acquisition engineer at Sioux Falls, SD based Bright Planet, told CBS News.'
Why not just force Twitter to change TOS to require sarcasm tag?"
Why not just force Twitter to change TOS to require sarcasm tag?"
Re:This should be easy. (Score:5, Insightful)
Just feed a bayesian analysis with postings from /. and use that to match against twitter. Should be able to get within 10%...
Some conspiracy insight - the government's quantum based AI big brother master computer of the NSA is up and running in order to completely surveil the internet in real time.
The problem is, it is nearly useless because it can't determine if someone is being sarcastic / snarky or not.
Sarcasm tags wouldn't work. (Score:4, Insightful)
Because people would instantly start using them sarcastically.
[sarcasm]It's like ten thousand spoons when all you need is a knife[/sarcasm]
Re:Detect Sarcasm???? (Score:5, Insightful)
Why would the Secret Service, in particular, want to tell sarcasm apart from other speech? Think about who they are.
They want to be able to distinguish sarcastic political speech, from sincere political speech. Of course both are protected speech.
Now, they might have a benign purpose, but from the description in TFA it doesn't seem so. After all, the administration would look pretty foolish if they tried to harass or jail someone for being sarcastic.
Re:This should be easy. (Score:4, Insightful)
Actually, what they could do would be to correlate English language postings with equivalent German language postings. As Germans are known world wide for their fun-loving sense of humor and sarcastic wits, the difference should yield accurate, non-snarky posts in English.