Tweak-the-Tweet

As many of you know, I’ve been doing a lot of research into social media analytics, especially in algorithms for text analysis of Twitter data. My focus has been on what the machine learning people call unsupervised learning. Why? Because I’ve come to the realization that tweets are an evolving language. They’re really a meta-language – a tweet could be about a web page, a blog, a picture, etc.

Tweets often aren’t complete sentences or other linguistic constructs as we know them. Twitter has @replies, hashtags, retweets, “follow friday”, trending topics, link shorteners, and a number of other new linguistic constructs that don’t appear in the natural human languages we use in everyday conversation.

There is now a project called Tweak-the-Tweet, TtT for short. TtT is a project of the University of Colorado at Boulder. Here’s the news release: CU Grad Student’s ‘Tweet’ Approach Streamlines Online Communications During Haiti Disaster.

As the story notes, Tweak-the-Tweet is helping Haiti relief efforts by providing standardized syntax for Twitter communications. I think this is very important. You can think of TtT as the Twitter equivalent of the telegraph and ham radio’s Morse code, or police and citizen band’s “10-codes.” It’s a way of conveying a lot of information in a small 140-character space. ReadWriteWeb covered the story here

Tweak the Tweet: New Twitter Hashtag Syntax for Sharing Information During Catastrophes

Here’s the main page for the project: HELPING HAITI: TWEAK the TWEET (TtT). You can follow them on Twitter and visit the project wiki. One of the projects we’ll be working on at this weekend’s CrisisCampPDX is Tweak-the-Tweet, so I’ll be posting an update next week. Meanwhile, I urge anyone who can to help out this worthwhile effort.