analyzing Twitter data using block-based programming

Features

Snap!Twitter is a extension for the block-based programming language Snap!. Particularly, it allows to conduct simple data analyses. For this purpose, it allows accessing Twitter data, which are a rich data source. Currently, the main features are:

  • access up to about 40 tweets per second
    (theoretical limit of the Twitter API, in most cases up to 5 tweets per second and browser window depending on JavaScript execution times)
  • use all meta data of the tweets (up to about 150 attributes) for analysis purposes
  • visualize data such as geo-coded tweets on interactive maps (based on leaflet.js)
  • get insight into data by analyzing them and showing the results as charts (based on plotly.js)
  • combine Snap!Twitter with other Snap! extensions, as it is compatible with most of them
  • extend Snap!Twitter by writing own blocks, as it is always possible in Snap!
  • modify Snap!Twitter blocks as they are written either in Snap! or JavaScript within Snap!

How to use

For getting a first impression on how Snap!Twitter looks like, you can just access our server (see below). The front-end runs completely in the browser, so you do not need to install anything. But you won't be able to run your analysis without first contacting us: As our server has limited performance, we need to keep track on how many classes are using it, in order to inform to try to prevent low performance of the analysis: At the moment, our Snap!Twitter server has proven to be sufficiently performant for being access by whole school classes. However, we expect analyses and tweet rates to go down when more than 50 students use it concurrently.

Hence, if you want to use it in school our try out the analysis, please contact us (see contact data at the footer).


Attention: Full access only after contacting us, authentication needed!

Instead of using our server, you might also set up an own server. The software is completely based on JavaScript and Node.js, it does not need an additional webserver as it includes one. The source is available on GitHub, so refer to the documentation there for instructions on how to set it up.

Contact

Andreas Grillenberger
Freie Universität Berlin
Didaktik der Informatik
Königin-Luise-Str. 24/26
14195 Berlin

E-Mail: andreas.grillenberger@fu-berlin.de

Source

Snap!Twitter is available completely open source and based on open source libraries. You can find all our sources on → GitHub.