That thanks to social networks we live more and more connected is something thatEsteban Morono doubt. However, it has also shown that there is much that we still do not know about how information flows on the Internet. That is why his research focuses on modeling human behavior on social networks. This physicist, who researches at the Carlos III University and the CSIC-UAM-UCM-UC3M Institute of Mathematical Sciences, devotes his attention to topics as diverse as random processes, financial mathematics, social networks, the study of fluids to micrometric scales, the growth of surfaces … In 2007 it won the IBM “Shared University Award” in the year for themodeling of the dissemination of information on social networks and its application to viral marketing. Today he is speaking at EmTechSpain, the emerging technologies conference organized by the MIT Institute in Spain.
Question. What have you discovered in your research about how information is transmitted on social networks?
Answer.Although the image we have of social networks is that information travels at high speed, what we have found is that, in most cases, viral information spreads very slowly due to human communication patterns. Humans communicate in “bursts”, which means a succession of calls, tweets or messages between two people is followed by periods in which there is no communication and that makes the speed of information much slower than expected . An example: the average response time for an email is one day, but we saw in the campaigns we carried out with IBM that the speed of propagation in email networks is 14 days … and even years! !
Q. Very slow … How does that affect the spread of rumors?
R.A lot of! Let them tell, for example, the Veuve Clicquot champagne brand! They have been fighting a hoax on social media for years (since 2003, no less) that the company gives away free champagne if you send a message through the company’s website. And the fact that information spreads so slowly means that some rumors take a long time to die, that is, they are very persistent.
Q. Does the microblogging platform Twitter have anything special?
R.Twitter is a fascinating platform, because unlike others it can be used to collect data and analyze it. We have worked on various problems of sensing opinion, urban mobility in large cities, etc., but the most recent thing we have done is answering the following question: can we put “sensors” on the network? That is, can we easily find a set of accounts / people on Twitter so that we learn better and faster about what information travels virally on the network (rumors, opinions, etc.)? What we have found is that, using the so-called “friendship paradox” in social networks – the fact that on average your friends in a social network have more friends than you – we can easily find people (sensors) who have a “position “on the web better than most users and they allow us to know in advance what is spreading virally.
P. Thanks to the Internet we are and will be more … How would the sentence end?
R.We will be more network! In my presentation at EmTech Spain I will show examples that the Redd and we are increasingly interdependent, that many of our behaviors come from the Internet. And vice versa: we can change our “neighbors” on the web. To the point that perhaps the object of study in many problems should change: it is the Web that has the nature to explain what is happening.
I’ll give you an example: during the strike on March 29, we collected all the tweets about the strike and in a video we showed how the information was exchanged and shared through retweets. Thus, we detected that there were two clear groups detected by search algorithms for online communities: one in favor and the other against the mobilization. And in the video you can clearly see the two groups fighting for a place in the “conversation” space on Twitter. The result seems to indicate that the network has a life of its own …
