TubePlot is an application for smartphones and tablets, designed specially for Facebook, which uses personal data to generate short audiovisual fictions. Using the method of probability calculation, the application generates customized video clips. The app designs video clips featuring in real-time the digital lives of Facebook-users analysing user’s personal data made available by Facebook via the Graph API. The clips generated by the application are similar to stories released by users on their Facebook accounts. However, there is one crucial difference: the clips produced by TubePlot are based on intrigues resulting from probability calculations generated on the basis of statistical and semantic analysis of the user’s personal data.
The application is based on a profiling system. The core of this profiling system is configured to infer a number of typical features of Facebook-users and their best friends, carried out on their tastes, preferences, desires and moods. The inferred data draw a raw canvas on which little scenarios are built by a data-driven plot-generator. As such, the application puts emphasis on the phenomenon of quantification and classification of human behaviour on social platforms. It uses probability calculations to define elements for the development of a new media concept that tells data-driven stories based on data of the user, friends, relations and connections on social Platforms. On Facebook our actions are not only published and stored, but also sorted, indexed and quantified: the frequency and time of our posts, the applications consulted, the posts or comments most discussed or shared, our likes and the words we have used frequently.