Project Ideas

Below are some ideas for Master’s theses, bachelor projects and other projects for students at the IT University of Copenhagen. Contact me (mcos@itu.dk) to get started.

  • Efficient estimation of a network’s polarization: I have developed a measure to estimate how polarized the political discourse is on Facebook and Twitter. However, the measure can only be calculated efficiently for small networks. In this project, you will explore new algorithms to improve the efficiency of the calculation.
  • Affective polarization on social media: having different opinions is not an issue, but it is when it causes people to talk to other people in aggressive way. This is affective polarization: the perception of disagreement as an enemy action. Can we estimate the level of affective polarization on a social network? How does it relate to opinion polarization?
  • Build Multilayer Networks from Graphic Novels: making a network out of a work of fiction is helpful to study literary theory and gender norms in fiction. Most works so far build networks with a single type of interaction, but multilayer and signed network could potentially be more useful. A good potential material could be Kentaro Miura’s Berserk, since it contains different arcs and characters changing allegiances.
  • Human mobility and development: Using a new dataset estimating the development level of places in developing countries, we can estimate the link between wealth and the place where one lives. Then we can investigate the effect of lockdowns on the livelihood of people: if human mobility gets much harder, are all people equally affected or is there a disproportionate negative impact for people living in low-income communities?
  • Study propagation on complex networks: How fast does a disease infect a social network? Is there a correlation between how two different diseases infect your friends? Can we tell how segregate messages are among communities ? Can we use these questions to figure out whether a message in a viral marketing campaign is passing through social connections that we are not able to observe? Can we answer these questions on networks that are evolving over time?
  • Multilayer Graph Association Rule Mining: Apply a newly developed multilayer graph mining algorithm to study the evolution of multilayer networks and perform multilayer link prediction: detect which new edges are likely to appear in the network, and in which layers. For instance: are Linkedin connections precursors of Facebook friendship? Is being connected on Twitter helping to create a social community in another platform? Is it better to take classes together at ITU or to be part of a study group, if we want to be friends?
  • Some other random keywords. If you find any of these intriguing, you can talk to me about them to see if we can carve a project out of them: Solving utilitarianism via pair ranking; Network similarity measures; Rise of the “shaming” subreddits; Multiplex Economic Complexity Index; Inflation of News; Central place theory and synthetic networks; The Oscar effect; Content-creation networks and meritocracy; Random geometric networks …