Social Networks on

Detecting bias in algorithms used to disseminate information in social networks

Sun, 26 Oct 2025 00:00:00 +0000

Authors: Vedran Sekara, Ivan Dotu, Manuel Cebrian, Esteban Moro, and Manuel Garcia−Herranz
Publication: PNAS Nexus, 2025, 4, pgaf291 [Journal | ArXiv]

Abstract: Social connections are conduits through which individuals communicate, information propagates, and diseases spread. Identifying individuals who are more likely to adopt ideas and spread them is essential in order to develop effective information campaigns, maximize the reach of resources, and fight epidemics. Consequently, a lot of work has focused on identifying influencers in social networks with various influence maximization algorithms being proposed. Based on extensive computer simulations on synthetic and 10 diverse real-world social networks we show that seeding information in social networks using state-of-the-art influence maximization methods creates information gaps. Our results show that these algorithms select influencers who do not disseminate information equitably, threatening to create an increasingly unequal society. To overcome this issue, we devise a multiobjective algorithm which both maximizes influence and information equity. Our results demonstrate it is possible to reduce vulnerability at a relatively low trade-off with respect to spread. This highlights that in our search for maximizing the spread of information we do not need to compromise on information equality.

Media

How Algorithmic outreach lead to information inequality, Health Medicine Network
Algorithmic Outreach Drives Growing Information Inequality, Scienmag

How to Hide One's Relationships from Link Prediction Algorithms

Tue, 22 Oct 2019 00:00:00 +0000

Authors: Marcin Waniek, Kai Zhou, Yevgeniy Vorobeychik, Esteban Moro, Tomasz P Michalak, Talal Rahwan
Journal: Scientific Reports volume 9, Article number: 12208 (2019) Link

Abstract: Our private connections can be exposed by link prediction algorithms. To date, this threat has only been addressed from the perspective of a central authority, completely neglecting the possibility that members of the social network can themselves mitigate such threats. We fill this gap by studying how an individual can rewire her own network neighborhood to hide her sensitive relationships. We prove that the optimization problem faced by such an individual is NP-complete, meaning that any attempt to identify an optimal way to hide one’s relationships is futile. Based on this, we shift our attention towards developing effective, albeit not optimal, heuristics that are readily-applicable by users of existing social media platforms to conceal any connections they deem sensitive. Our empirical evaluation reveals that it is more beneficial to focus on “unfriending” carefully-chosen individuals rather than befriending new ones. In fact, by avoiding communication with just 5 individuals, it is possible for one to hide some of her relationships in a massive, real-life telecommunication network, consisting of 829,725 phone calls between 248,763 individuals. Our analysis also shows that link prediction algorithms are more susceptible to manipulation in smaller and denser networks. Evaluating the error vs. attack tolerance of link prediction algorithms reveals that rewiring connections randomly may end up exposing one’s sensitive relationships, highlighting the importance of the strategic aspect. In an age where personal relationships continue to leave digital traces, our results empower the general public to proactively protect their private relationships.

¿Cuánta gente vas a acabar conociendo en toda tu vida?

Fri, 06 Sep 2019 00:00:00 +0000

Article (in Spanish) in the spanish newspaper El País about how many people we will know in our life. This is a summary of our recent research on how humans create/destroy relationships and how narrow and small is our world even throughout a life of encounters, relationships, work, etc.

¿Cuánta gente vas a acabar conociendo en toda tu vida? Muy poca

En el mundo hay más de 7.500 millones de personas. En la Unión Europea hay más de 500 millones. En América Latina, más de 600. En Madrid viven 3,1 millones; en Buenos Aires, 2,8; en Barcelona, 1,6. Sin embargo, en 60 años de vida adulta nuestra limitada capacidad cognitiva solo nos permitirá cruzarnos con tanta gente como hay en Becerril de la Sierra (Madrid), Alcudia de Crespins (Valencia) o Cacabelos (León): unos 5.000. Las redes sociales, la tecnología y la facilidad de movimiento dan la sensación de que tenemos el mundo en las manos. Pero es solo una sensación. La cifra de gente con la que mantendremos un contacto que puede ir desde un par de conversaciones –un fontanero– hasta nuestra pareja es reducida. Ese grupo incluye toda la gente susceptible de convivir contigo, de aparecer en tu vida o de influir en tus decisiones: familiares, profesores, colegas, amigos y toda nuestra experiencia directa. Nuestra capacidad de comprender íntimamente las motivaciones de personas distintas es ridícula. Apenas existe.

Quizá hay quien piense que no puede ser. Es cierto que hay seres especiales capaces de conocer a mucha más gente: quizá 10.000 o más. O gente con oficios que lleven a más contactos: solo en la agenda de un periodista puede haber miles de contactos. El número es de hecho una mediana. “Pero lo que está claro es que esa cifra no es 1 millón, ni siquiera 50.000”, dice Esteban Moro, investigador del MIT Media Lab, de la Universidad Carlos III y autor de numerosos trabajos sobre redes.

The dynamic character of our networked society

Sat, 02 Feb 2019 00:00:00 +0000

We live in a networked society and our actions, opinions, behaviors are affected and can affect other people. Understanding such social networked structures is one of the key challenges in our attempt to decode human behavior and its impact in our society. Although human interactions are dynamical by nature, most of our understanding relies in static representations of those social networks. However, social interactions are rarely static. Very often the networks evolve by means of processes that happen at diverse time scales, like link decay/formation, group formation, etc. Our research in the last years have been to develop dynamical models of social networks that account for all those processes at different temporal scales. Understanding how networks evolve has allowed us to reveal new behavior patterns hidden in those social dynamics, but also their effect in societal problems like information diffusion, viral marketing or social mobilization.

For example, using massive datasets of viral marketing campaigns and mobile phone calls by 20 million people, we were the first group to identify the slowing down of information diffusion in social networks due to the inhomogeneous (bursty) activity of humans. This finding explained also why marketing techniques based on spreading of information (viral marketing) are sometimes unsuccessful to achieve expected reach and coverage in reasonable time. But also, it warns us about the potential risk of using network strategies in situations of critical mobilization, a study we published in PNAS: social mobilization can be very fast, but it is unreliable. Part of this research, and specially the understanding of viral marketing campaigns, was recognized by IBM with “2007 Shared University Research” award.

At a different temporal scale, humans create and decay human interactions every day. The unavailability of large longitudinal databases about human interactions prevented the understanding of what are the main strategies behind our social dynamics. In collaboration with Telefónica and accessing the largest dataset (almost two years) used in this kind of research, we were the first group to identify what is the strategy that individuals use to create and decay social relationships while maintaining a constant amount of time/attention to those relationships. We found the universal result that people can be classified as keepers or explorers depending on whether they create less or more relationships for a given capacity to maintain a certain number of relationships. Our finding of those universal strategies have been corroborated by other groups in email or human mobility, for example.

We have worked extensively in this line of research to find how other temporal patterns of social interactions can predict long-term successful relations or the importance of daily rhythms in social relationships. Finally, we have know a collaboration with Facebook to use their data to understand how strong ties are created in our society whether is it possible to predict their creation.

Some recent papers:

Temporal patterns behind the strength of persistent ties
Henry Navarro, Giovanna Miritello, Arturo Canales, Esteban Moro
EPJ Data Science 6 97 (2017)
PDF Journal Altmetric: 35
Twitter session analytics: Profiling users\textquoteright short-term behavioral changes
Farshad Kooti, Esteban Moro, Kristina Lerman
In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (2016)
PDF Proceedings Altmetric: 5
Channel-Specific Daily Patterns in Mobile Phone Communication
Talayeh Aledavood, Eduardo Lopez, Sam G B Roberts, Felix Reed-Tsochas, Esteban Moro, Robin I M Dunbar, Jari Saramaki
In Proceedings of ECCS 2014 (2016)
PDF Chapter
Time allocation in social networks: correlation between social structure and human communication dynamics
Giovanna Miritello, Ruben Lara, Esteban Moro
Preprint 175– (2013)
PDF arXiv Altmetric: 7
Limits of social mobilization
Alex Rutherford, Manuel Cebrian, Sohan D’souza, Esteban Moro, Alex Pentland, Iyad Rahwan
Proceedings Of The National Academy Of Sciences Of The United States Of America 110 6281–6286 (2013)
PDF Journal Altmetric: 55
Limited communication capacity unveils strategies for human interaction.
Giovanna Miritello, Ruben Lara, Manuel Cebrian, Esteban Moro
Scientific Reports 3 1950 (2013)
PDF Journal Altmetric: 37

Talks:

Dynamics in Complex Networks: Analysing Real-world data my talk (video) at the Workshop on Complex Network Mining and Analysis. Transparencies can also be found here

Media:

Growing old in Twitter

Fri, 14 Dec 2018 00:00:00 +0000

I started using Twitter more than 10 years ago (!). I open an account in this social network in 2008 and although I was not using it too much for the first year, I become a frequent user after that. It has helped me to get news, information both for my personal and professional interests. But not only that, Twitter has been also the data source for our research, that helped us to investigate the relationship between human behavior in the social platform and paramount problems in our society as information propagation, unemployment, disaster damage, political opinion. As we keep on working on those subjects we have also recently extended our research to other problems like health, or climate change

Most of the research in human behavior is constrained either by time or by population covered, so we can’t have both. Large longitudinal databases, extending tens of years, are relatively small in the number of participants or users, while data from millions of users is usually obtained for a very short period of time (months or years). One of the good things about growing old in those social networks is that we are starting to see tens of years of data to analyze.

Here I want to analyze by Twitter activity during those last 10 years. First thing is to download all our account activity, something that is explain in the How to download and view your Twitter archive help page at Twitter. Basically:

Connect to your Account Settings at Twitter.
On the left sidebar you will see a link to Your Twitter data.
At this step you will probably have to confirm your password, but on the bottom of the next page you have a link to Request data of your Twitter account.
When the data is ready to download you will receive an notification at your email with the a link to download it.
The data comes as a series of JSON files. The file tweets.js contains all tweets, retweets and metions, but it comes with a window.YTD.tweet.part0 = header at the beggining. Remove it to make it a readable JSON file.

Let’s load the tweets

require(jsonlite)
tweets <- jsonlite::fromJSON("tweets.js")

The table contains many fields, including the tweet id (id), timestamp when it was created created_at, if it is a reply to a status in_reply_to_status_id or a user in_reply_to_user_id.

More or less active?

The first thing we can investigate is if my behavior in Twitter has changed in these 10 years. My feeling is that people spend less time in the platform when we get older. One reason is that, compared to 2009, it is really difficult to keep tract of what is happening in the platform. I also have less and less time. But it is true that twitter has changed their app to engage users more with the converstation, so that might counterbalance it.

To analyze it, let’s add the formated timestamp to the dataset

tweets$timestamp <- as.POSIXct(tweets$created_at,format="%a %b %d %H:%M:%S %z %Y")

and plot the number of tweets by month.

require(ggplot2)
require(zoo)
ggplot(tweets,aes(x=as.yearmon(timestamp))) + 
  geom_bar(binwidth=.09)+scale_x_yearmon() + 
  theme_bw() + 
  ylab("Number of tweets per month") + xlab("Time")

As we can see, the most active years were from 2011 to 2015 (around 100 tweets per month). From then on I am tweeting less, corroborating my feeling that I spent less time in the platform (at least tweeting :) )

Tweeting or retweeting more?

Have I changed the way I use Twitter? Our research in Twitter sessions and social networks using mobile phone data shows that because of our limited atention and cognitive capacities people tend to perform simpler tasks with time and age. For example we found that in long sessions in Twitter (two or more hours), users start composing less messages (which require more effort) and use more retweets or mentions (replies) within the session, that require less effort.

Let’s see what happened in ten years of data. We classify tweets as composed, mention or retweets using the fields in the dataset.

tweets$class <- "normal"
tweets$class[!is.na(tweets$in_reply_to_status_id)] <- "mention"
tweets$class[!is.na(tweets$retweeted_status_id)] <- "RT"
tweets$class[grep("RT @",tweets$full_text)] <- "RT"

and show the fraction of tweets per month in each class

require(ggplot2)
require(zoo)
ggplot(tweets,aes(x=as.yearmon(timestamp), fill=class)) + 
  geom_bar(position="fill",binwidth=.09)+
  scale_x_yearmon() + theme_bw() + 
  ylab("Fraction of tweets of each class") + xlab("Time")

Similar to our research for Twitter sessions, I can see that I compose less original tweets with time: while in 2010 almost 50% of my tweets were composed, now only 20% are original and more than 50% of the tweets in my account are retweets.

Tweeting about what?

Finally, let’s see what I tweeted about. Although we could probably do much elaborated analysis, a simple wordcloud will do here. We clean up the text of the tweets (including mentions and retweets) and produce a wordcloud

require(tm)
require(wordcloud2)
texts <- tweets$full_text
#cleanup remove mentions and url
texts <- tolower(texts)
texts <- gsub('\\b+rt', '', texts)
texts <- gsub("@\\S+", "", texts)
texts <- gsub('http\\S+\\s*', '', texts)
texts <- gsub('[[:punct:]]', '', texts) 
texts <- removeWords(texts, stopwords("english")) #get rid of stopwords in english
texts <- removeWords(texts, stopwords("spanish")) #get rid of stopwords in spanish
corpus.texts.all <- Corpus(VectorSource(texts))
dtm <- TermDocumentMatrix(corpus.texts.all)
m <- as.matrix(dtm)
v <- sort(rowSums(m),decreasing=TRUE)
d <- data.frame(word = names(v),freq=v)
wordcloud2(d[d$freq>20,],fontFamily="Loma",rotateRatio = 0)

As you can see, the word I used more is thanks (“gracias” in spanish), together with other related to my field of research (networks, data, social, etc.).

Conclusion

So, yes I am growing older in Twitter and the way I use the platform is different. I probably spend more time reading that engaging in new conversations, new tweets or creating hashtags for events or conferences. This exercise proves that not only Twitter is a good platform to understand timely events like elections, sports, natural disasters or unemployment, but also to understand how people change in the course of a lifetime (or at least tens of years) with and outside the platform.

And as my wordcloud shows, I am thankful for those 10 years of sharing the platform with friends, colleagues and other familiar strangers!

Talk at NICO "The lifetime of strong ties in social networks"

Wed, 11 Oct 2017 00:00:00 +0000

Video of my talk “The lifetime of strong ties in social networks” at the Northwestern Institute of Complex Systems (NICO) in Evaston, Chicago, Oct 2017. What a great experience giving a talk at NICO!

Temporal patterns behind the strength of persistent ties

Wed, 28 Jun 2017 12:54:18 +0000

Authors: Henry Navarro, Giovanna Miritello, Arturo Canales, Esteban Moro
Journal: EPJ Data Science (2017) 6:31 LINK

Abstract: Social networks are made out of strong and weak ties having very different structural and dynamical properties. But what features of human interaction build a strong tie? Here we approach this question from a practical way by finding what are the properties of social interactions that make ties more persistent and thus stronger to maintain social interactions in the future. Using a large longitudinal mobile phone database we build a predictive model of tie persistence based on intensity, intimacy, structural and temporal patterns of social interaction. While our results confirm that structural (embeddedness) and intensity (number of calls) features are correlated with tie persistence, temporal features of communication events are better and more efficient predictors for tie persistence. Specifically, although communication within ties is always bursty we find that ties that are more bursty than the average are more likely to decay, signaling that tie strength is not only reflected in the intensity or topology of the network, but also on how individuals distribute time or attention across their relationships. We also found that stable relationships have and require a constant rhythm and if communication is halted for more than 8 times the previous communication frequency, most likely the tie will decay. Our results not only are important to understand the strength of social relationships but also to unveil the entanglement between the different temporal scales in networks, from microscopic tie burstiness and rhythm to macroscopic network evolution.

Media

Important relationships are not bursty Read more about the implications of our work to detect long-lasting relationships.

El romance entre Twitter y la ciencia

Mon, 30 Jan 2017 00:00:00 +0000

Nice article (in Spanish) in Yorokobu magazine about the use of Twitter in science, highlighting our work on the use of social media for rapid assessment of natural disaster management

El romance entre Twitter y la ciencia

«Twitter y las otras redes sociales están entre los más grandes archivos de actividad humana existentes y, al contrario que con las encuestas, ahora tenemos la posibilidad de analizar millones de mensajes, opiniones e interacciones de personas en diferentes contextos como la política, economía o el ocio», explica el matemático Esteban Moro, investigador en la Carlos III de Madrid. «Esto nos ha permitido estudiar fenómenos que antes eran imposibles de medir, como la discusión en tiempo real sobre temas de política o los movimientos de millones de personas en un país».

Searching for someone

Fri, 19 Feb 2016 11:52:48 +0000

From the “Small World Experiment” to the “Red Balloon Challenge,” and beyond

We live in a small world, right? But the cost and fragility of navigating it could harm any potential strategy to leverage the power of social networks. Read this fascinating story of the research, experiments, and failures in the quest for using social networks to search information/people:

[Excerpt of the article] Our ability to search social networks for people and information is fundamental to our success. We use our personal connections to look for new job opportunities, to seek advice about what products to buy, to match with romantic partners, to find a good physician, to identify business partners, and so on.

Despite living in a world populated by seven billion people, we are able to navigate our contacts efficiently, only needing a handful of personal introductions before finding the answer to our question, or the person we are seeking. How does this come to be? In folk culture, the answer to this question is that we live in a “small world.” The catch-phrase was coined in 1929 by the visionary author Frigyes Karinthy in his Chain-Links essay, where these ideas are put forward for the first time.

Keep reading at MIT Media Lab Medium

Daily rhythms in mobile telephone communication

Tue, 13 Oct 2015 13:23:36 +0000

Authors:Talayeh Aledavood , Eduardo López, Sam G. B. Roberts, Felix Reed-Tsochas, Esteban Moro, Robin I. M. Dunbar, Jari Saramäki
Journal: PLoS ONE 10(9), e0138098 (2015) LINK

Abstract: Circadian rhythms are known to be important drivers of human activity and the recent availability of electronic records of human behaviour has provided fine-grained data of temporal patterns of activity on a large scale. Further, questionnaire studies have identified important individual differences in circadian rhythms, with people broadly categorised into morning-like or evening-like individuals. However, little is known about the social aspects of these circadian rhythms, or how they vary across individuals. In this study we use a unique 18-month dataset that combines mobile phone calls and questionnaire data to examine individual differences in the daily rhythms of mobile phone activity. We demonstrate clear individual differences in daily patterns of phone calls, and show that these individual differences are persistent despite a high degree of turnover in the individuals’ social networks. Further, women’s calls were longer than men’s calls, especially during the evening and at night, and these calls were typically focused on a small number of emotionally intense relationships. These results demonstrate that individual differences in circadian rhythms are not just related to broad patterns of morningness and eveningness, but have a strong social component, in directing phone calls to specific individuals at specific times of day.

From Seconds to Months: multi-scale dynamics of mobile telephone calls

Thu, 07 May 2015 12:39:43 +0000

Authors:Jari Saramaki, Esteban Moro
Journal: Eur. Phys. J. B (2015) 88: 164 LINK | **arXiv

Abstract: Big Data on electronic records of social interactions allow approaching human behaviour and sociality from a quantitative point of view with unforeseen statistical power. Mobile telephone Call Detail Records (CDRs), automatically collected by telecom operators for billing purposes, have proven especially fruitful for understanding one-to-one communication patterns as well as the dynamics of social networks that are reflected in such patterns. We present an overview of empirical results on the multi-scale dynamics of social dynamics and networks inferred from mobile telephone calls. We begin with the shortest timescales and fastest dynamics, such as burstiness of call sequences between individuals, and “zoom out” towards longer temporal and larger structural scales, from temporal motifs formed by correlated calls between multiple individuals to long-term dynamics of social groups. We conclude this overview with a future outlook.

Social media fingerprints of unemployment

Thu, 07 May 2015 12:39:43 +0000

Authors:Alejandro Llorente, Manuel García-Herránz, Manuel Cebrián and Esteban Moro
Journal: PLoS ONE 10(5): e0128692 (2014) LINK

Summary: Publicly available social media data can be used to quantify deviations from typical patterns of behavior and uncover how these deviations signal the socio-economical status of regions. Using data from geolocalized Twitter messages, we find that unemployment is correlated with technology adoption, daily activity, diversity in mobility patterns and correctness in communication style. These behavioral metrics serve to build simple, interpretable, and cost-effective socio-economical predictors from these novel digital datasets. Our extensive investigation allows us not only to build accurate behavioral models of how unemployment impacts diverse geographical areas, but also to assessing the relevance and uniqueness of previously reported social media datasets to understand economical development.

Abstract Recent wide-spread adoption of electronic and pervasive technologies has enabled the study of human behavior at an unprecedented level, uncovering universal patterns underlying human activity, mobility, and inter-personal communication. In the present work, we investigate whether deviations from these universal patterns may reveal information about the socio-economical status of geographical regions. We quantify the extent to which deviations in diurnal rhythm, mobility patterns, and communication styles across regions relate to their unemployment incidence. For this we examine a country-scale publicly articulated social media dataset, where we quantify individual behavioral features from over 145 million geo-located messages distributed among more than 340 different Spanish economic regions, inferred by computing communities of cohesive mobility fluxes. We find that regions exhibiting more diverse mobility fluxes, earlier diurnal rhythms, and more correct grammatical styles display lower unemployment rates. As a result, we provide a simple model able to produce accurate, easily interpretable reconstruction of regional unemployment incidence from their social-media digital fingerprints alone. Our results show that cost-effective economical indicators can be built based on publicly-available social media datasets.

Media

See the video of thousands of trips in Spain used to characterize the mobility between municipalities in Spain

Press

Comunidades de partidarios en redes sociales: estudio de las elecciones catalanas de 2010 y 2012

Tue, 22 Apr 2014 12:48:19 +0000

Authors: Esteban Moro
Book: Cotarelo, R. & Olmeda, J.A. (Comps.) (forthcoming). La democracia del siglo XXI. Política, medios de comunicación, internet y redes sociales. Actas de las II Jornadas españolas de ciberpolítica, 28 de mayo de 2013. Madrid: Centro de Estudios Políticos y Constitucionales. [pdf]

Abstract: En los últimos años hemos asistido a un incremento notable de eventos de carácter político y/o social que han sido promovidos (cuando no originados) a través de medios de comunicación electrónicos. En particular, el uso cada vez más frecuente de plataformas de sociabilidad electrónicas como Facebook o Twitter para compartir opinión o información sobre temas de actualidad, política o sociales ha hecho que estás plataformas pasen a ser herramientas imprescindibles dentro de la comunicación política y también los nuevos canales en los que se comparten ideas, se busca información o se organizan campañas dentro del contexto político.

Uno de los ámbitos más estudiados ha sido el de las elecciones, en el que la discusión política está acotada tanto en ámbito como en tiempo y donde, por tanto, se produce un mayor posicionamiento político en las redes sociales de los usuarios, de los partidos, asociaciones, etc. El objetivo de esta contribución es estudiar este fenómeno de formación de grupos partidarios (partisanos) en el flujo de información en Twitter en un contexto político más variado en las que concurren una mayor diversidad opciones políticas (ver video). Asimismo pretendemos estudiar la caracterización, estabilidad de dichas comunidades y, en particular, la correlación entre sus propiedades y la estimación de voto. El hecho de disponer de datos con dos años de separación entre dos elecciones nos permite comprobar también la adecuación e invariabilidad de la metodología del análisis de redes sociales en dos situaciones diferentes.

Using Friends as Sensors to Detect Global-Scale Contagious Outbreaks

Wed, 09 Apr 2014 13:37:55 +0000

Authors: Manuel Garcia-Herranz, Esteban Moro, Manuel Cebrian, Nicholas A. Christakis and James H. Fowler
Journal: PLoS ONE 9(4): e92413 (2014) LINK

SummaryRecent research has focused on the monitoring of global–scale online data for improved detection of epidemics, mood patterns, movements in the stock market political revolutions, box-office revenues, consumer behaviour and many other important phenomena. However, privacy considerations and the sheer scale of data available online are quickly making global monitoring infeasible, and existing methods do not take full advantage of local network structure to identify key nodes for monitoring. Here, we develop a model of the contagious spread of information in a global-scale, publicly- articulated social network and show that a simple method can yield not just early detection, but advance warning of contagious outbreaks. In this method, we randomly choose a small fraction of nodes in the network and then we randomly choose a friend of each node to include in a group for local monitoring. Using six months of data from most of the full Twittersphere, we show that this friend group is more central in the network and it helps us to detect viral outbreaks of the use of novel hashtags about 7 days earlier than we could with an equal-sized randomly chosen group. Moreover, the method actually works better than expected due to network structure alone because highly central actors are both more active and exhibit increased diversity in the information they transmit to others. These results suggest that local monitoring is not just more efficient, but also more effective, and it may be applied to monitor contagious processes in global–scale networks.

Press coverage:

  * The science of going viral, [The Chronicle of Higher Education](http://chronicle.com/blogs/percolator/the-science-of-going-viral/34145)
  * How to Use Twitter Friends as Sensors to Detect Disease Outbreaks, [Technology Review MIT](http://www.technologyreview.com/view/508091/how-to-use-twitter-friends-as-sensors-to-detect-disease-outbreaks/)
  * Going viral: How 'social contagion' begins and escalates, [Bloomberg Businessweek](http://investing.businessweek.com/research/markets/news/article.asp?docKey=600-201404101313M2______EUPR_____7af500000170ad87_3600-1)
  * Nice summary of our work at the [Robert Wood Johnson Foundation](http://www.rwjf.org/en/research-publications/find-rwjf-research/2014/04/using-friends-as-sensors-to-detect-global-scale-contagious-outbr.html) webpage
  * How 'social contagion' begins and escalates, [PhysOrg](http://phys.org/news/2014-04-social-contagion-escalates.html)
  * System detects global trends in social networks two months in advance, [ScienceDaily](http://www.sciencedaily.com/releases/2014/04/140428094211.htm)
  * Tus tuits predicen el futuro, [Huffington Post](http://www.huffingtonpost.es/2014/04/29/twitter-predice-el-futuro_n_5231224.html)
  * Predicting contagious outbreaks using your most popular friends [FastCompany.com](http://www.fastcoexist.com/3029058/predicting-contagious-outbreaks-using-your-most-popular-friends)
  * The Case for Journeying to the Center of Our Social Networks, [RWJF.com](http://www.rwjf.org/en/blogs/pioneering-ideas/2014/05/the_case_for_journey.html?cid=xtw_pioneer)

** **

Performance of Social Network Sensors During Hurricane Sandy

Thu, 13 Feb 2014 13:57:24 +0000

Authors: Yury Kryvasheyeu, Haohui Chen, Esteban Moro, Pascal Van Hentenryck, Manuel Cebrian
Journal: PLoS ONE 10(2): e0117288 (2015) LINK

Summary Information flow during catastrophic events is a critical aspect of disaster management. Modern communication platforms, in particular online social networks, provide an opportunity to study such flow, and a mean to derive early-warning sensors, improving emergency preparedness and response. Performance of the social networks sensor method, based on topological and behavioural properties derived from the “friendship paradox”, is studied here for over 50 million Twitter messages posted before, during, and after Hurricane Sandy. We find that differences in user’s network centrality effectively translate into moderate awareness advantage (up to 26 hours); and that geo-location of users within or outside of the hurricane-affected area plays significant role in determining the scale of such advantage. Emotional response appears to be universal regardless of the position in the network topology, and displays characteristic, easily detectable patterns, opening a possibility of implementing a simple “sentiment sensing” technique to detect and locate disasters.

Time allocation in social networks: correlation between social structure and human communication dynamics

Fri, 17 May 2013 07:47:26 +0000

Authors: Giovanna Miritello, Rubén Lara, and Esteban Moro
Book: “Temporal Networks”, Springer, 2013. Series: Understanding Complex Systems. Holme, Petter; Saramaki, Jari (Eds.) [PDF]((http://arxiv.org/pdf/1305.3865v1.pdf)

Summary Recent research has shown the deep impact of the dynamics of human interactions (or temporal social networks) on the spreading of information, opinion formation, etc. In general, the bursty nature of human interactions lowers the interaction between people to the extent that both the speed and reach of information diffusion are diminished. Using a large database of 20 million users of mobile phone calls we show evidence this effect is not homogeneous in the social network but in fact, there is a large correlation between this effect and the social topological structure around a given individual. In particular, we show that social relations of hubs in a network are relatively weaker from the dynamical point than those that are poorer connected in the information diffusion process. Our results show the importance of the temporal patterns of communication when analyzing and modeling dynamical process on social networks.

Limits of social mobilization

Mon, 01 Apr 2013 20:37:19 +0000

Authors: Alex Rutherford, Manuel Cebrian, Sohan Dsouza, Esteban Moro, Alex Pentland, and Iyad Rahwan
Journal: PNAS 110 (16), 6281-6286 (2013) LINK

Abstract: The Internet and social media have enabled the mobilization of large crowds to achieve time-critical feats, ranging from mapping crises in real time, to organizing mass rallies, to conducting search-and-rescue operations over large geographies. Despite significant success, selection bias may lead to inflated expectations of the efficacy of social mobilization for these tasks. What are the limits of social mobilization, and how reliable is it in operating at these limits? We build on recent results on the spatiotemporal structure of social and information networks to elucidate the constraints they pose on social mobilization. We use the DARPA Network Challenge as our working scenario, in which social media were used to locate 10 balloons across the United States. We conduct high-resolution simulations for referral-based crowdsourcing and obtain a statistical characterization of the population recruited, geography covered, and time to completion. Our results demonstrate that the outcome is plausible without the presence of mass media but lies at the limit of what time-critical social mobilization can achieve. Success relies critically on highly connected individuals willing to mobilize people in distant locations, overcoming the local trapping of diffusion in highly dense areas. However, even under these highly favorable conditions, the risk of unsuccessful search remains significant. These findings have implications for the design of better incentive schemes for social mobilization. They also call for caution in estimating the reliability of this capability.

Press:

How social media mobilizes society [LiveScience]
La movilización por redes sociales puede ser rápida, pero tiene límites (EFE, El Diario Vasco, La Información, Muy Interesante) [link,link2]
Researchers root out the limits of social mobilization [Phys.org]
Usar las redes sociales para movilizar tiene límites [UC3M, ABC.es, Finanzas.com EFE, AgenciaSinc, Madri+d]
Using social networks for mobilization has its limits [UC3M, ScienceNewsLine]
¿Son las Redes sociales una buena herramienta de movilización social en situaciones críticas ? (audio) [RTVE]
Las movilizaciones en redes sociales son muy eficientes, si están dirigidas [Tendencias21]
Facebook passes research test for quick responses, collaboration [NBCNEWS.com]
Social media has limited mobilisation power [ABC Science]
Las movilizaciones en las redes sociales son rápidas, pero con un alto riesgo de fracaso [RDI Press]
Researchers look into the use of social networks in emergency communications [Continuity Central]
Un experimento pone a prueba a las redes sociales [La Razón]
Masdar Institute Researchers Quantify the ‘Reliability’ of Social Media for Time-Critical Mobilization [Zawya]
MIT researchers quantify ‘reliability’ of social media for mass mobilisation [The Gulf Today]
El poder y la flaqueza de las redes sociales [Conec]

Time as a limited resource: Communication Strategy in Mobile Phone Networks

Mon, 14 Jan 2013 08:52:59 +0000

Authors: Giovanna Miritello, Esteban Moro, Rubén Lara, Rocío Martínez-López, Sam G. B. Roberts, Robin I. M. Dunbar
Journal: Social Networks 35, 89 (2013) LINK | arXiv

Abstract: We used a large database of 9 billion calls from 20 million mobile users to examine the relationships between aggregated time spent on the phone, personal network size, tiestrength and the way in which users distributed their limited time across their network (disparity). Compared to those with smaller networks, those with large networks did not devote proportionally more time to communication and had on average weaker ties (as measured by time spent communicating). Further, there were not substantially different levels of disparity between individuals, in that mobile users tend to distribute their time very unevenly across their network, with a large proportion of calls going to a small number of individuals. Together, these results suggest that there are time constraints which limit tie strength in large personal networks, and that even high levels of mobile communication do not fundamentally alter the disparity of time allocation across networks.

Temporal network of information diffusion in Twitter

Mon, 29 Oct 2012 21:58:29 +0000

Millions of tweets, retweets and mentions are exchanged in Twitter everyday about very different subjects, events, opinions, etc. While aggregating this data over a time window might help to understand some properties of those processes in online social networks, the speed of information diffusion around particular time-bound events requires a temporal analysis of them. To show that (and with the help of the Text & Opinion Mining Group at IIC) we collected all tweets (750k) of the vibrant conversation around the disputed subject of the general strike of March 29th in Spain. The data spans 10 days from 03/27 to 04/04 and using the RTs related to the general strike between twitter accounts we build up the following temporal network of information diffusion in Twitter.

Day/night human rhythms are clearly seen, and there is an increase of activity in the evening/night before March 29th, which ended in the burst of RTs during that day. Moreover, using community-finding algorithms over the static (weighted) network of RTs we could assign each twitter account to one of the communities found. Analyzing the text of tweets within those communities we found the nature of the biggest groups: one is in favor of the economic motivations behind the strike, the other is not. Those communities fight close to dominate information propagation in Twitter even some days after the strike.

This video highlights the importance of temporal networks in the analysis of information diffusion in online social networks.

Technical details: the video was done using the amazing igraph package in R and encoded using ffmpeg. Thanks to everyone that contributes to those open-source projects for their work.

Edit (11/9/2012): I have post a tutorial on how to make this kind of visualizations here. Spread the word!

Predicting Human Preferences Using the Block Structure of Complex Social Networks

Wed, 12 Sep 2012 12:19:43 +0000

Authors: Roger Guimerà, Alejandro Llorente, Esteban Moro y Marta Sales-Pardo
Journal: PLoS ONE 7, e44620 (2012) LINK

Abstract: With ever-increasing available data, predicting individuals’ preferences and helping them locate the most relevant information has become a pressing need. Understanding and predicting preferences is also important from a fundamental point of view, as part of what has been called a “new” computational social science. Here, we propose a novel approach based on stochastic block models, which have been developed by sociologists as plausible models of complex networks of social interactions. Our model is in the spirit of predicting individuals’ preferences based on the preferences of others but, rather than fitting a particular model, we rely on a Bayesian approach that samples over the ensemble of all possible models. We show that our approach is considerably more accurate than leading recommender algorithms, with major relative improvements between 38% and 99% over industry-level algorithms. Besides, our approach sheds light on decision-making processes by identifying groups of individuals that have consistently similar preferences, and enabling the analysis of the characteristics of those groups.

Branching dynamics of viral information spreading

Thu, 03 Nov 2011 09:51:44 +0000

Authors: José Luis Iribarren and Esteban Moro
Journal: Physical Review E 84, 046116 (2011) LINK | arXiv

Abstract: Despite its importance for rumors or innovations propagation, peer-to-peer collaboration, social networking, or marketing, the dynamics of information spreading is not well understood. Since the diffusion depends on the heterogeneous patterns of human behavior and is driven by the participants’ decisions, its propagation dynamics shows surprising properties not explained by traditional epidemic or contagion models. Here we present a detailed analysis of our study of real viral marketing campaigns where tracking the propagation of a controlled message allowed us to analyze the structure and dynamics of a diffusion graph involving over 31 000 individuals. We found that information spreading displays a non-Markovian branching dynamics that can be modeled by a two-step Bellman-Harris branching process that generalizes the static models known in the literature and incorporates the high variability of human behavior. It explains accurately all the features of information propagation under the “tipping point” and can be used for prediction and management of viral information spreading processes.

Social Features of Online Networks: The Strength of Intermediary Ties in Online Social Media

Thu, 21 Jul 2011 09:06:07 +0000

Authors: P. A. Grabowicz, J. J. Ramasco, E. Moro, J. P. Pujol and V. M. Eguiluz
Journal: PLoS ONE 7(1): e29358 (2012). LINK

Abstract: An increasing fraction of today’s social interactions occur using online social media as communication channels. Recent worldwide events, such as social movements in Spain or revolts in the Middle East, highlight their capacity to boost people’s coordination. Online networks display in general a rich internal structure where users can choose among different types and intensity of interactions. Despite this, there are still open questions regarding the social value of online interactions. For example, the existence of users with millions of online friends sheds doubts on the relevance of these relations. In this work, we focus on Twitter, one of the most popular online social networks, and find that the network formed by the basic type of connections is organized in groups. The activity of the users conforms to the landscape determined by such groups. Furthermore, Twitter’s distinction between different types of interactions allows us to establish a parallelism between online and offline social networks: personal interactions are more likely to occur on internal links to the groups (the weakness of strong ties); events transmitting new information go preferentially through links connecting different groups (the strength of weak ties) or even more through links connecting to users belonging to several groups that act as brokers (the strength of intermediary ties).

Twitter y Política: Información, Opinión y ¿Predicción?

Fri, 17 Jun 2011 10:56:12 +0000

Authors: M. Luz Congosto, Montse Fernández y Esteban Moro
Journal: Cuadernos de Comunicación Evoca 4 (2011). LINK

Abstract: Estamos a las puertas de una nueva manera de medir la opinión política: mediante la conversación en Red, que permite no sólo conocer el feedback de los ciudadanos a la política, sino también la imagen de los políticos que se refleja en la Red, la dinámica de opinión en comunidades o grupos y el efecto de los diferentes medios de comunicación en dicha conversación. En este sentido, Twitter es una de las más interesantes fuentes públicas de datos en tiempo real, por la que fluye información muy valiosa que puede impulsar el avance en el estudio de la demanda social relacionada con la política.

Complex Dynamics of Human Interactions, September 14th 2011

Mon, 28 Mar 2011 14:51:50 +0000

We (together with Kimmo Kaski, Aalto University) are organizing the ECCS'11 Satellite conference “Complex Dynamics of Human Interactions” to be held at Vienna, September 14th.

You can find more info at http://www.complexdynamics.org

“It’s not enough to have a map of the structure. It is crucial to understand the dynamics of a process”, L. Barábasi

Scope The nature of human interaction has undergone a substantial change in the past years and the change does not seem to be over. Technologies like email, smart-phones, social networks like Facebook or broadcast technologies like Twitter transform the way people keep in touch and new trends of communication appear: individuals are continuously connected with each other, social activities are commonly shared by groups of people and people do not need to be geographically close to stay connected.

The high availability of digital data about human activity given by these communication channels and their high detail has provided unprecedented understanding of the nature of humans interactions, that affect the very definition of social relationships, hubs, communities and their role on society. Particular important is the role that human dynamics has in processes that happen concurrently with the dynamics of interaction, like information/disease epidemics in social networks, opinion dynamics, coordination, etc.

The aim of this meeting is to explore the dynamical structure of social and communication networks and the role of the human complex dynamics in realistic processes like information spreading, personal recommendation or “word-of-mouth”, etc.

Specific topics of interest are (but not only):

  * High Frequency analysis of communication and social networks
  * Causality and correlation in human communication patterns
  * Reality Mining, Face-to-Face interactions
  * Modeling dynamics of human interactions
  * Applications to viral marketing, infection spreading, opinion dynamics.

Affinity Paths and information diffusion in social networks

Tue, 08 Feb 2011 09:42:47 +0000

Authors: José Luis Iribarren and Esteban Moro
Journal: Social Networks 33, 134-142 (2011). LINK | arXiv

Abstract: Widespread interest in the diffusion of information through social networks has produced a large number of Social Dynamics models. A majority of them use theoretical hypothesis to explain their diffusion mechanisms while the few empirically based ones average out their measures over many messages of different contents. Our empirical research tracking the step-by-step email propagation of an invariable viral marketing message delves into the content impact and has discovered new and striking features. The topology and dynamics of the propagation cascades display patterns not inherited from the email networks carrying the message. Their disconnected, low transitivity, tree-like cascades present positive correlation between their nodes probability to forward the message and the average number of neighbors they target and show increased participants’ involvement as the propagation paths length grows. Such patterns not described before, nor replicated by any of the existing models of information diffusion, can be explained if participants make their pass-along decisions based uniquely on local knowledge of their network neighbors affinity with the message content. We prove the plausibility of such mechanism through a stylized, agent-based model that replicates the Affinity Paths observed in real information diffusion cascades.

Dynamical strength of social ties in information spreading

Thu, 25 Nov 2010 10:19:44 +0000

Authors: Giovanna Miritello, Esteban Moro y Rubén Lara
Journal: Physical Review E (Rapid Comm) 83, 045102 (2011). LINK | arXiv

Abstract: We investigate the temporal patterns of human communication and its influence on the spreading of information in social networks. The analysis of mobile phone calls of 20 million people in one country shows that human communication is bursty and happens in group conversations. These features have opposite effects in information reach: while bursts hinder propagation at large scales, conversations favor local rapid cascades. To explain these phenomena we define the dynamical strength of social ties, a quantity that encompasses both the topological and temporal patterns of human communication.

Relationship mining

Mon, 11 Jan 2010 09:16:09 +0000

Each day trillions of emails, phone calls, comments on blogs, twitter messages, exchanges in online social networks, etc. are done. Not only the number of communications has increased, but also each of these transactions leaves a digital trace that can be recorded to reconstruct our high-frequency human activity. It is not only the amount and variety of data that is recorded what is important. Also its high-frequency character and its comprehensive nature have allowed researchers, companies and agencies to investigate individual and group dynamics at an unprecedented level of detail and applied them to client modeling, organizational analysis or epidemic spreading [1].

However, for technical or privacy reasons only the existence but not of the content of those exchanges is known. Thus we can quantify the intensity and frequency of the interaction but not its type. For decades, social science has measured relationships between individuals in the currency of tie strength, introduced by Granovetter [1]. Weak ties (loose acquaintances) can help to disseminate ideas and/or innovations between different groups, help to find a job or new information; while strong ties (family, trusted friends) hold together organizations and social groups and can affect emotional health. Despite its success to explain these phenomena, tie strength of human relationships is vaguely defined in most large-scale social empirical work. Specifically, relationships are generally quantified by the intensity or duration of communication, although they are known to have significant drawbacks as tie strength predictor [3,4]. Multiplexity, rhythm and depth of the communication seem to be better predictors of tie strength than intensity [4]. Incorporating those metrics in the data mining of online communication might improve the definition of relationships between individuals and in turn transform our understanding of individual dynamics and its impact in our lives, organizations and society [5]. The challenge is to unveil social relationships in social media and not just mere interactions between individuals, which in general over-represent the real structure of a social group [6] (see figure). And this is of paramount importance to understand the propagation of ideas, opinions, commercial messages, etc. in social networks, since most links declared in social networks might be meaningless from a relationship point of view.

[caption id=“attachment_441” align=“aligncenter” width=“500” caption=“Undressing the social network: considering all e-mail interactions in a academic social network (left) yields to a highly dense and connected social network, while strong interactions (based on the individual relative frequency of communication) render the social group sparser and disconnected”] [/caption]

References

  1. D. Lazer et al. _Computational Social Science_, Science **323**, 721 (2009)
  2. M. S. Granovetter, _The Strength of Weak Ties_, The American Journal of Sociology **78(6)**, 1360 (1973)
  3. P. V. Marsden, and K. E. Campbell _Measuring Tie Strength_ Social Forces **63(2)**, 482 (1990).
  4. E. Gilbert and K. Karahalios, _Predicting Tie Strength with Social Media_, presented in CHI 2009.
  5. C. T. Butts, _Revisting the Foundations of Network Analysis_, Science **325**, 414 (2009)
  6. B. A. Huberman, D. M. Romero, and F. Wu, _Social networks that matter_, First Monday **14(1)** (2009).

Note: This article appears in the Catalog of the exhibition “Culturas del Cambio: Átomos Sociales y Vidas Electrónicas” in the Center _Arts Santa Mónica. _Thanks to Josep Perelló for his kind invitation to contribute

Impact of Human Activity Patterns on the Dynamics of Information Diffusion

Tue, 04 Aug 2009 07:59:19 +0000

Authors: J. L. Iribarren and E. Moro
Journal: Physical Review Letters 103, 038702 (2009) LINK arXiv

Abstract: We study the impact of human activity patterns on information diffusion. To this end we ran a viral email experiment involving 31183 individuals in which we were able to track a speciﬁc piece of information through the social network. We found that, contrary to traditional models, information travels at an unexpectedly slow pace. By using a branching model which accurately describes the experiment, we show that the large heterogeneity found in the response time is responsible for the slow dynamics of information at the collective level. Given the generality of our result, we discuss the important implications of this ﬁnding while modeling human dynamical collective phenomena.

Press coverage:

‘Infectious’ people spread memes across the web, New Scientist (12/08/09)
Email hoaxes are like viruses, The Inquirer (10/08/09)
The flow of viral video, ABC News (8/08/09)
New model for social marketing campaigns details why some information ‘goes viral’, PhysOrg (6/08/09)
Los perezosos frenan los rumores en Internet, ABC.es (14/8/09)

Publish and... perish

Mon, 12 Jan 2009 08:08:00 +0000

In the old days, research quality was measured by the number of papers you published. Publishing was a hard process and only few scientists were able to publish several papers per year. However, with the bloom of new journals, the appearance of electronic editorial process, and the specialization of research fields, the number of publications per year has grow exponentially during the last decades. Thus publishing is not longer a good measure of the quality of research. As an example if this I recently attended a talk by Sid Redner in which he showed the following data extracted from the Physical Review citation data of 353000 papers:

  * Only 11 papers got more than 1000 citations
  * 245000 got less than 10 citations
  * 100000 got one or none citations

For me it is amazing to see that roughly 1/3 of the papers published got almost no citations at all. It is a publish and perish process in which 1/3 of the papers are lost.

The situation is similar to what has been found recently in the music industry. Out of 13 million sons available to buy online, 10 million of them have never been bought. As Will Page put it:

  * Only 20% of the tracks in their sample are 'active', that is to say they sold at least one copy, and hence, 80% of the tracks sold nothing
  * 80% of the revenue came from around 3% of the active tracks
  * Only 4 tracks sold more than 100000 copies

This led Will Page to question the Long-Tail theory by Chris Anderson which states that the market share of low demanded items can be bigger than that of best-sellers. To put it in mathematical terms, the mass of the distribution in the tail of can be bigger than the mass around the peak of the distribution. This happens mostly with Pareto-law distributions and thus the name “long-tail”. But Will Page’s data seems to suggest that there is not even such a tail and planning your business in the long tail is risky: if you center your business plan in trying to sell the tail of the distribution, most probably you won’t succeed. As Andrew Bud (Executive Chairman from Mblox) put it: “in this tail, you starve”

The same can happen to a journal if it lives in the long-tail: what is the fraction of perishable papers a editor of a journal is willing to accept? What is the “citation model” the journal is intending to have? We all know about the impact factor of a journal, which is only giving us information about (mostly) regular-cited papers. A better information will be also the zero-index of a journal (or a researcher), i.e. the fraction of papers that never get cited at all. An idea I am working on recently… Stay tuned

Information diffusion epidemics in social networks

Wed, 06 Jun 2007 07:37:32 +0000

Authors: José Luis Iribarren, Esteban Moro
Journal: Preprint 0706.0641 arXiv

Abstract: The dynamics of information dissemination in social networks is of paramount importance in processes such as rumors or fads propagation, spread of product innovations or “word-of-mouth” communications. Due to the difficulty in tracking a specific information when it is transmitted by people, most understanding of information spreading in social networks comes from models or indirect measurements. Here we present an integrated experimental and theoretical framework to understand and quantitatively predict how and when information spreads over social networks. Using data collected in Viral Marketing campaigns that reached over 31,000 individuals in eleven European markets, we show the large degree of variability of the participants’ actions, despite them being confronted with the common task of receiving and forwarding the same piece of information. This have a profound effect on information diffusion: Firstly, most of the transmission takes place due to super-spreading events which would be considered extraordinary in population-average models. Secondly, due to the different way individuals schedule information transmission we observe a slowing down of the spreading of information in social networks that happens in logarithmic time. Quantitative description of the experiments is possible through an stochastic branching process which corroborates the importance of heterogeneity. Since high variability of both the intensity and frequency of human responses are found in many other activities, our findings are pertinent to many other human driven diffusion processes like rumors, fads, innovations or news which has important consequences for organizations management, communications, marketing or electronic social communities.

Press:

This research has been awarded by IBM with a “Shared University Research” 2007 grant. Read more about that (in spanish):