Tweetbeat helps beat reporters keep track of the buzz on their beats via Twitter.
What it does
TweetBeat is a web application that allows a user to search within a specific community of handles on Twitter. The project was inspired by Fuego, a Nieman Lab project, that is constantly updated and displays the most popular links being shared by influential users in the journalism innovation community. TweetBeat lets users create their own communities to explore. Users give TweetBeat five influencer handles, and TweetBeat track those five handles in addition to all the handles that these original five follow. The technology then returns popular hashtags and keywords this extended network is currently tweeting about. TweetBeat also analyzes the network and finds other influential handles.
How it works
Users enter five similar Twitter handles on the TweetBeat homepage, the technology then uses the Twitter REST API to collect the array of relevant handles. The algorithm builds a large list of these “friends” and then chooses the top ranked friends by the number of times they appear in the compiled list. TweetBeat then takes the top handles from this list and requests their user timeline to collect relatively recent tweets from each handle. After all the tweets are collected into a large array, TweetBeat breaks them down into components including keywords, hashtags and top links. Once the tweets are broken down and analyzed in the Python backend, these three areas along with the top ranked users are set to the HTML templates to display to the user. Whenever a new search is conducted, TweetBeat rebuilds the network from the new input, and all the new data is collected again via the Twitter API.
- Build an application that runs in the background to continue to enhance the user network and its accuracy and relevancy.
- Allow a user to adjust the timeframe for a search instead of most recent tweets only.
- Include more comprehensive information than the four categories currently displayed, such as specific tweets, statistics about the results and more.
- Cache the data in a database so that user sessions could be saved and revisited at a later time without having to change the timeframe parameters or the Twitter handles entered.
Student Team: Megha Singh Das, Philip House, Denise Lu
Faculty Guidance: Larry Birnbaum, Rich Gordon