Problem Description

Social media have become an integral part of everyday life. It is therefore not surprising to see an increasing number of conversations taking place on social networking sites such as Twitter. In comparison to Facebook, Twitter is a platform oriented toward individual statements, whereby the most diverse points of view lead to conversation. Twitter itself provides only limited and partial data via its own interface. However, this data can be tracked and reconstructed into a large whole using suitable methodology and the interface provided by Twitter.

Therefore, the rich reconstruction and extraction of information based on sporadic and partial data offered via the Twitter interface is problematic.


In order to provide long-term data and in particular conversations, i.e., the thematic and self-relating arrangement of individual statements, and the information beyond this, such as user behavior, overarching thematic links, the following points must be met:

  1. review of existing methods and data sets from the generally accepted literature
  2. testing and evaluation of the knowledge gained from point 1
  3. implementation of the knowledge gained from point 2
  4. further development of the methods obtained from point 3

Overall, methods for reconstructing and extracting Twitter data as well as its contextual processing are to be identified, investigated and advanced.

Interest in social media, data analysis and programming


Marc Feger
Raum 25.12.O2.46 ·