The Tweetographer captures data on upcoming events thematically within a specific geographical area or a given category, and then makes a list available to its users.
What if the data flowing in from Twitter could be turned into a guide to what’s happening in your neighbourhood? This is the ambition of two computer science students at the University of Cincinnati in the United States who have developed the Tweetographer. This web application, based on data-mining, collects tweets from a large number of users within a given geographical area, automatically extracts information on parties, concerts, sports events, etc., and produces a real-time compilation of these happenings.
Real-time listings according to various parameters
The Tweetographer starts from the principle that Twitter is now the main platform for Internet users to post online what they are planning to do or hope to do. The application focuses on ascertaining the nature of the current and future events mentioned and assessing how popular they are. The results are then shown in lists according to different parameters – type, date and place. The app can also be set to show only specific categories of results, in order to prevent it from being overwhelmed by huge streams of data, especially in big cities.
Potentially diverse uses of the app
The students also had to develop a thesaurus of multiple Twitter synonyms for all kinds of words since, in the 140 million tweets posted every day, users type many different variations of the same word. During testing, the software demonstrated its effectiveness, immediately displaying onscreen tweets posted just seconds beforehand. The two students hope to make further improvements to the tool going forward, and to find other useful applications for it, such as predicting the outcome of elections or compiling product reviews and information for brands and companies.