Public defence in Computer Science, M.Sc. (Tech) Sara Heydari

Title of the thesis: Dynamic yet persistent: investigating digital traces of human behaviour

Doctoral student: Sara Heydari
Opponent: Professor Jukka-Pekka Onnela, Harvard T.H. Chan School of Public Health, USA
Custos: Professor Jari Saramäki, Aalto University School of Science, Department of Computer Science

In today’s world, every small action we take - whether making a call, browsing the web, or driving on a highway - leaves behind a digital trace. Analysing these traces gives the opportunity to understand human behaviour on a scale that was impossible before. This thesis uses a variety of digital traces to unravel the attributes of our social networks, as well as the population-level patterns of commuting and travelling.

This thesis investigates the properties of personal social networks by analysing millions of ego-networks inferred based on traces of communication on different mediums (such as mobile phone calls, sms, email and internet forums). The results show that we tend to have individual preferences in how to shape our personal social networks and that the shapes of these social networks are stable, even if the people we are in touch with can change vastly over time.

To further examine these persistent social networks, a model was built that connects the structure of personal social networks to people’s communication strategies. Fitting the model to the communication data reveals how each person balances her tendency to deepen existing relationships with the urge to form new ties.

Additionally, this thesis uses high-resolution digital communication logs and computational methods to validate the predictions of sociological theories that predate the digital age. In the 1980’s, Scott Feld hypothesised that multiplex and monoplex ties (social relationships that span multiple or single social contexts) have different roles in connecting us as a society. This thesis demonstrates that even without any information about the context of social interactions, it is possible to identify multiplex social ties based on the timings of the communications. The multiplex and monoplex ties identified with this method contribute to social network connectivity in distinct ways, which aligns with Feld’s theory.

Moreover, this thesis also shows the potential of digital traces for real-time estimation of country-wide mobility flows even amidst large distortions in mobility patterns, such as those we experienced during the COVID-19 pandemic.

In summary, this thesis uses digital traces and computational methods to advance our understanding of human behaviour and bridges the novel field of computational social science with the rich tradition of classical social science.

Key words: Social Networks, Ego-networks, Social Signatures, Human Mobility

Thesis available for public display 10 days prior to the defence at: https://aaltodoc.aalto.fi/doc_public/eonly/riiputus/ 

Contact information:

Doctoral theses at the School of Science: https://aaltodoc.aalto.fi/handle/123456789/52 

Zoom Quick Guide: https://www.aalto.fi/en/services/zoom-quick-guide