The Gateway to Algorithmic and Automated Trading

CEU Professor contributes to study that shows technology spreads like disease

First Published 23rd October 2014

A dataset recording Skype-spreading dynamics gives evidence of media and social interactions in innovation adoption.

Professor Janos Kertesz, Centre for Network Science, CEU

Budapest - Professor Janos Kertesz from the Centre for Network Science at Central European University contributed to a collegial study on innovation dissemination with researchers from French Ecole Normale Supérieure de Lyon and Aalto University (Helsinki).

The scientists investigated the spreading of online innovations using big data and came to the conclusion that social adoption of innovation is similar to the pattern of epidemic contagion. Although these mechanisms have been identified by quantitative theories, their role and relative importance are not entirely understood, since empirical verification has been so far hindered by the lack of appropriate data.

Through the analysis of detailed records from Skype, the world's largest voice over Internet protocol, the researchers have brought to light evidence that empirically support the assumptions behind models of social contagion. The study conclusions are that the theoretical framework for disease spreading can be adopted to this complex social contagion process by incorporating effects of social pressure and of the media. Diffusion of innovation can be interpreted as a social spreading phenomena governed by the impact of media and social interactions. The results of the study highlights three mains findings:

  • the probability of spontaneous service adoption is constant - meaning people who adopt a service based on pervasive media or advertisements.

  • the probability of adoption via social influence is linearly proportional to the fraction of adopting neighbors - meaning the likelihood of you joining the service is directly proportional to the number of friends you have who use it.

  • the probability of service termination is time-invariant and independent of the behavior of peers.
By implementing the detected diffusion mechanisms into a dynamical agent-based model, the research team was able to emulate the adoption dynamics of the service in several countries worldwide. The findings of this first of its kind approach enabled them to make medium- term predictions of service adoption and disclose dependencies between the dynamics of innovation spreading and the socioeconomic development of a country.
The more tech savvy and open a society is, the higher the likelihood of innovation spreading more rapidly.

The full study is available in Journal of the Royal Society Interface: