Merkmalsanalyse bei plattformübergreifender Benutzeridentifikation in sozialen Medien

The usage of social media has been propagated into the daily life of the modern age. However, not everyone uses social media in the right way and for a useful purpose. Abusing social media for criminal and terrorist activities has been a problem since they emerged. As one individual might be operating different profiles for either benign or malicious activities, sometimes there exists the requirement to prove that the same individual uses two or more profiles of different social media platforms. This case arises in open-source investigations where internal access logs are not available for the investigators. Previous work has demonstrated that this task is feasible by using proper machine learning techniques. However, it has remained unclear, which extracted features in this machine learning task had been the most important ones to recognize the users. The goal of this work is to find out an answer to this research question. 

Although Python is the preferred programming language, the utilized programming language can vary based on the student's knowledge and preferences. 


  1. Zhou et al.,  "Cross-Platform Identification of Anonymous Identical Users in Multiple Social Media Networks",  2016
  2. Liu et al.,  "HYDRA: Large-scale Social Identity Linkage via Heterogeneous Behavior Modeling", 2014
  3. Liu et al., "User Identification Cross Multiple Social Media Platform with Revised Input Output Network Embedding Framework", 2019
  4. Veiga and Eickhoff.,  "A Cross-Platform Collection of Social Network Profiles", 2016.
  5. Cross-OSN: