Science may be difficult, but it definitely doesn’t have to be ugly. These images, from a new exhibit at the British Library, show how beautiful scientific data can be.
The exhibit features classic illustrations dating to 1603, including John Snow’s map of London’s SoHo that’s credited with revealing a contaminated water pump as the source of a 1854 cholera outbreak. There also are beautiful modern visualizations of data from satellites and gene sequencers. The exhibit, Beautiful Science: Picturing Data, Inspiring Insight, runs through May 26.
“Some games (both single-player and massive multiplayer) already include complex economic systems as part of their design mechanic. It would be easy to embed real-world, real-time financial data (such as commodity prices) within those games. Companies could track how players react to the data, then aggregate the reactions to predict real-world economic events accordingly. And, as massive multiplayer games with rich, dynamic economies become increasingly popular, opportunities to learn from player behavior will be enhanced accordingly.” (Edery, 2006)
Lewis and colleagues analyzed the donation and recruitment activity of more than 1 million members of the Save Darfur Cause between May 2007 and January 2010. About 80 percent of the members had been recruited by other members and about 20 percent had joined independently.
Of these 1 million-plus members, 99.76 percent never donated any money and 72.19 percent never recruited anyone else.
The Save Darfur Cause on Facebook raised only about $100,000. While the average donation amounts were similar to more traditional fundraising methods ($29.06), the donation rate was much smaller: 0.24 percent. Compare that to mail solicitations which typically yield donation rates of 2 to 8 percent. The larger Save Darfur campaign, the researchers note, raised more than $1 million through direct-mail contributions in fiscal year 2008 alone.
Interestingly, those that had joined the Facebook cause independently were both more likely to donate and to recruit.
Social and financial contributions, though rare on both counts, also tended to go hand-in-hand. Those individuals that did recruit were nearly four times as likely as non-recruiters to donate. And donors were more than twice as likely as non-donors to recruit.
The data contained no demographic information on the cause's members. Nor, the researchers write, could they estimate "the personal significance of [the joining] gesture to participants or the symbolic impact of the movement to onlookers.
"It is possible," they add, "that the individuals in our data set contributed to Save Darfur in other meaningful but unobserved ways."
Still, Lewis and colleagues believe the study gives some valuable insights into collective action in a digital age.
In most social and information systems the activity of agents generates rapidly evolving time-varying networks. The temporal variation in networks' connectivity patterns and the ongoing dynamic processes are usually coupled in ways that still challenge our mathematical or computational modelling. Here we analyse a mobile call dataset and find a simple statistical law that characterize the temporal evolution of users' egocentric networks. We encode this observation in a reinforcement process defining a time-varying network model that exhibits the emergence of strong and weak ties. We study the effect of time-varying and heterogeneous interactions on the classic rumour spreading model in both synthetic, and real-world networks. We observe that strong ties severely inhibit information diffusion by confining the spreading process among agents with recurrent communication patterns. This provides the counterintuitive evidence that strong ties may have a negative role in the spreading of information across networks.
The number of followers you have and the exact wording matter less than you think. What makes a difference is having the right message for the right people.
"Influence" doesn't necessarily mean what you think it does. In the age of the social-media celebrity, a glut of Twitter followers or particularly pugnacious sampling of pithy updates are often the hallmarks of an influencer. But new research suggests that influence is situational at best: as people compete for the attention of the broader online ecosystem, the relevance of your message to the existing conversation of those around you trumps any innate "power" a person may have.
.... According to co-author Vespignani, having millions of followers does not denote an important message. Rather, the messages with the most immediate relevance tend to have a higher probability of resonating within a certain network than others. Think of it as "survival of the fittest" for information: those tweets that capture the most attention, whether related to a major geopolitical or news event or a particular interest, are likely to persist longer. This competition sounds bad, but it's generally good for messages in general: thousands of tweets about Japan's 2011 earthquake or the ongoing conflict in Syria don't cancel each other out, but help refocus the attention of the wider Twitter audience on those issues, which in turn provides an added lift to individual messages over other off-topic ones.
Social networks are visual in nature. Visualization techniques have been applied in social analysis since the field began. We aim to develop interactive visual analytic tools for complex social networks.
In this agile marketing infographic, possible influencer touchpoints are connected by what we call “cultivation bands” of tactical flow. Each band is broken down into initial contact point with a potentially engaged follower in social channels – who is then generated as a qualified lead, nurtured with engaging content and then madesocial media advocate – before the user becomes the most positive aspect of all digital marketing user phases…a contributor.
PLOS ONE: an inclusive, peer-reviewed, open-access resource from the PUBLIC LIBRARY OF SCIENCE. Reports of well-performed scientific studies from all disciplines freely available to the whole world.
Social influence is the process by which individuals adapt their opinion, revise their beliefs, or change their behavior as a result of social interactions with other people. In our strongly interconnected society, social influence plays a prominent role in many self-organized phenomena such as herding in cultural markets, the spread of ideas and innovations, and the amplification of fears during epidemics. Yet, the mechanisms of opinion formation remain poorly understood, and existing physics-based models lack systematic empirical validation.
Here, we report two controlled experiments showing how participants answering factual questions revise their initial judgments after being exposed to the opinion and confidence level of others. Based on the observation of 59 experimental subjects exposed to peer-opinion for 15 different items, we draw an influence map that describes the strength of peer influence during interactions. A simple process model derived from our observations demonstrates how opinions in a group of interacting people can converge or split over repeated interactions.
In particular, we identify two major attractors of opinion: (i) the expert effect, induced by the presence of a highly confident individual in the group, and (ii) the majority effect, caused by the presence of a critical mass of lay people sharing similar opinions. Additional simulations reveal the existence of a tipping point at which one attractor will dominate over the other, driving collective opinion in a given direction.
These findings have implications for understanding the mechanisms of public opinion formation and managing conflicting situations in which self-confident and better informed minorities challenge the views of a large uninformed majority.
Data scientists trace how the most-viewed video in YouTube history spread across the Internet
When South Korean pop star Psy released his “Gangnam Style” video in 2012 it spread like wildfire. Researchers at Indiana University Bloomington tracked the spreading meme by following how Twitter users shared the video with friends and strangers alike. By the time 200 tweets had linked to the video among the subset of Twitter users studied, “Gangnam Style” had already reached 86 different communities of users (blue nodes). After 3,000 tweets the meme had spread to nearly 1,000 different communities (green). “Gangnam Style” soon became the most-viewed video in YouTube history; by late 2013, the video had amassed more than 1.8 billion views.
How does network structure affect diffusion? Recent studies suggest that the answer depends on the type of contagion. Complex contagions, unlike infectious diseases (simple contagions), are affected by social reinforcement and homophily.
Our method aims to discover viral memes. To label viral memes, we rank all memes in our dataset based on numbers of tweets or adopters, and define a percentile threshold. A threshold of θT or θUmeans that a meme is deemed viral if it is mentioned in more tweets than θT% of the memes, or adopted by more users than θU% of the memes, respectively. All the features are computed based on the first 50 tweets for each hashtag h. Two baselines are set up for comparison. Random guessselects nviral memes at random, where nviral is the number of viral memes in the actual data.Community-blind prediction employs the same learning algorithm as ours but without the community-based features. We compute both precision and recall for evaluation; the former measures the proportion of predicted viral memes that are actually viral in the real data, and the latter quantifies how many of the viral memes are correctly predicted. Our community-based prediction excels in both precision and recall, indicating that communities are helpful in capturing viral memes (Fig. 5). For example, when detecting the most viral memes by users (θU = 90), our method is about seven times as precise as random guess and over three times as precise as prediction without community features. We achieve a recall over 350% better than random guess and over 200% better than community-blind prediction. Similar results are obtained using different community detection methods or different types of social network links (see SI).
ORCA (Organizational, Relationship, and Contact Analyzer) started by linking people who had been arrested together—the most objective way a record shows that people have, at the very least, been at the same place at the same time. From there, it categorized those who had admitted a gang affiliation. And then, based on social links, it gave the others a numerical probability of a particular affiliation. ORCA further analyzed clustered nodes within the network to identify groups and subgroups—a crew occupying a street corner, for example. By zeroing in on people connected across many groups and subgroups, ORCA singled out the most influential ones.
Describing a social network based on a particular type of human social interaction, say, Facebook, is conceptually simple: a set of nodes representing the people involved in such a network, linked by their Facebook connections. But, what kind of network structure would one have if all modes of social interactions between the same people are taken into account and if one mode of interaction can influence another? Here, the notion of a “multiplex” network becomes necessary. Indeed, the scientific interest in multiplex networks has recently seen a surge. However, a fundamental scientific language that can be used consistently and broadly across the many disciplines that are involved in complex systems research was still missing. This absence is a major obstacle to further progress in this topical area of current interest. In this paper, we develop such a language, employing the concept of tensors that is widely used to describe a multitude of degrees of freedom associated with a single entity.
Our tensorial formalism provides a unified framework that makes it possible to describe both traditional “monoplex” (i.e., single-type links) and multiplex networks. Each type of interaction between the nodes is described by a single-layer network. The different modes of interaction are then described by different layers of networks. But, a node from one layer can be linked to another node in any other layer, leading to “cross talks” between the layers. High-dimensional tensors naturally capture such multidimensional patterns of connectivity. Having first developed a rigorous tensorial definition of such multilayer structures, we have also used it to generalize the many important diagnostic concepts previously known only to traditional monoplex networks, including degree centrality, clustering coefficients, and modularity.
We think that the conceptual simplicity and the fundamental rigor of our formalism will power the further development of our understanding of multiplex networks.
The VC Partner social graph helps quantify and illustrate the level of connectivity among VC partners and their ties to successful (and not successful companies).
The Venture Capital Partner Social Graph (alpha)
Given the importance of networks in VC performance (aka network centrality), we've begun building a VC social graph which helps to quantify and illustrate the level of connectivity among VC partners and their ties to successful (and not successful companies). The end goal is to use network centrality and other measures to help LPs (and even firms themselves) understand who are the partners that create value and who might be the proverbial "dead wood". This is not dissimilar to our Investor Mosaic algorithms but applied to individual VC partners.
This extract of the VC Partner Social Graph looks at a set of elite VCs - specifically 26 individual VCs who sit on the Boards of 5 or more Tech IPO Pipeline companies we identified at the end of 2013. The VC Partner Social Graph also uses various algorithms to asses betweenness, reach and influence of individuals. And for fun, we applied some facial recognition algorithms to understand which of these 26 investors seemed happiest based on their pictures.
Social networks readily transmit information, albeit with less than perfect fidelity. We present a large-scale measurement of this imperfect information copying mechanism by examining the dissemination and evolution of thousands of memes, collectively replicated hundreds of millions of times in the online social network Facebook. The information undergoes an evolutionary process that exhibits several regularities. A meme's mutation rate characterizes the population distribution of its variants, in accordance with the Yule process. Variants further apart in the diffusion cascade have greater edit distance, as would be expected in an iterative, imperfect replication process. Some text sequences can confer a replicative advantage; these sequences are abundant and transfer "laterally" between different memes. Subpopulations of the social network can preferentially transmit a specific variant of a meme if the variant matches their beliefs or culture. Understanding the mechanism driving change in diffusing information has important implications for how we interpret and harness the information that reaches us through our social networks.
Information Evolution in Social Networks Lada A. Adamic, Thomas M. Lento, Eytan Adar, Pauline C. Ng
As for modes of transmission, violence seems to spread not just via human connections but also via media, even when the violence portrayed is fictional. One study found a correlation between a child’s television-viewing habits and aggressive behavior 15 years later. “Men who were high TV-violence viewers in childhood,” the authors wrote, “were convicted of crimes at over three times the rate of other men” . And in 2010, a major review of research involving more than 130,000 participants reported a causal relationship between exposure to violence in video games and aggressive behavior .
The wide adoption of social media has increased the competition among ideas for our finite attention. We employ a parsimonious agent-based model to study whether such a competition may affect the popularity of different memes, the diversity of information we are exposed to, and the fading of our collective interests for specific topics. Agents share messages on a social network but can only pay attention to a portion of the information they receive. In the emerging dynamics of information diffusion, a few memes go viral while most do not. The predictions of our model are consistent with empirical data from Twitter, a popular microblogging platform. Surprisingly, we can explain the massive heterogeneity in the popularity and persistence of memes as deriving from a combination of the competition for our limited attention and the structure of the social network, without the need to assume different intrinsic values among ideas.
Here we outline a number of empirical findings that motivate both our question and the main assumptions behind our model. We then describe the proposed agent-based toy model of meme diffusion and compare its predictions with the empirical data. Finally we show that the social network structure and our finite attention are both key ingredients of the diffusion model, as their removal leads to results inconsistent with the empirical data.
We first explore the competition among memes. In particular, we test the hypothesis that the attention of a user is somewhat independent from the overall diversity of information discussed in a given period. Let us quantify the breadth of attention of a user through Shannon entropy S = −Σi f(i) log f(i) where f(i) is the proportion of tweets generated by the user about meme i. Given a user who has posted n messages, her entropy can be as small as 0, if all of her posts are about the same meme; or as large as log n if she has posted a message about each of n different memes. We can measure the diversity of the information available in the system analogously, defining f(i) as the proportion of tweets about meme i across all users. Note that these entropy-based measures are subject to the limits of our operational definition of a meme; finer or coarser definitions would yield different values.
New research outlines the six types of communities on the social network and what that means for communication
Fil Menczer, a professor at the University of Indiana Bloomington School of Informatics and Computing, has researched the potential applications of this type of analysis for years. Menczer’s research touches on every aspect of Twitter’s role as a mirror for human communities, like examining the relationship between social data and the stock market, the spread of infectious diseases and how political campaigns manipulate data to spread misleading information. In a 2012 paper on the spread of memes on Twitter, Menczer and his team sought to demystify how information spreads on unrelated topics, yielding similar network structures to those uncovered by Pew.
One of the major lessons of network analysis, both Pew and Menczer emphasize, is that the Twitter commons hasn’t necessarily made society as democratic as techno-utopians would have you believe. Twitter isn’t a wide-open space, free of boundaries or obstacles: It’s a "mirror," as Menczer says, for the social structures of the real world.
“One of the presumptions about the rise of social media is that it’s changed everything,” says Himelboim. “In fact, if you look at the broadcast networks and brand clusters (two archetypes described by Pew), big, important and powerful institutions that wield tremendous influence offline still do on the Internet. This is really a reality check against those louder voices who claim the world has somehow been transformed."
“It makes you wonder about polarization in political discourse: Is this something that social media is responsible for?” asks Menczer. “Is more polarization easier because of social media, or are we observing what was already there with new technology? Or, even simpler: Would our discourse be better if Twitter and Facebook just didn’t exist?”
(Phys.org) —The question of how an economic system should be structured in order to best promote fairness and equality is one of the most debated subjects of all time. By approaching the complexities of this question from the field of network science, researchers from MIT and other institutions have ...
In their study, the researchers constructed a model in which individuals can earn income in two ways: by producing content or by distributing the content produced by others. A system in which more income is earned by production than by distribution is labeled as meritocratic, while one in which more income is earned by distribution is called topocratic. Importantly, the income earned by distribution depends not on what an individual produces but rather on an individual's position in the network.
Using this simple model, the researchers showed that the connectivity of the network determines whether the income is earned in a meritocratic or topocratic manner: densely connected networks are more meritocratic, while sparsely connected networks are more topocratic.
The difference makes sense, since individuals in densely connected networks can sell what they produce directly to others, and therefore do not need to share much of their proceedings with middlemen. On the other hand, in sparsely connected networks, individuals do not have direct connections with buyers and must rely on middlemen to help them connect.
In tracking 68 of these initiatives for one year after their inception, we discovered some striking predictors of change agents’ success. The short story is that their personal networks—their relationships with colleagues—were critical. More specifically, we found that:
1. Change agents who were central in the organization’s informal network had a clear advantage, regardless of their position in the formal hierarchy.
2. People who bridged disconnected groups and individuals were more effective at implementing dramatic reforms, while those with cohesive networks were better at instituting minor changes.
3. Being close to “fence-sitters,” who were ambivalent about a change, was always beneficial. But close relationships with resisters were a double-edged sword: Such ties helped change agents push through minor initiatives but hindered major change attempts.
We examine how firms can create word-of-mouth peer influence and social contagion by designing viral features into their products and marketing campaigns. To econometrically identify the effectiveness of different viral features in creating social contagion, we designed and conducted a randomized field experiment involving the 1.4 million friends of 9,687 experimental users on Facebook.com. We find that viral features generate econometrically identifiable peer influence and social contagion effects. More surprisingly, we find that passive-broadcast viral features generate a 246% increase in peer influence and social contagion, whereas adding active-personalized viral features generate only an additional 98% increase. Although active-personalized viral messages are more effective in encouraging adoption per message and are correlated with more user engagement and sustained product use, passive-broadcast messaging is used more often, generating more total peer adoption in the network. Our work provides a model for how randomized trials can identify peer influence in social networks.
Researchers are forecasting which memes will spread far and wide
What makes a meme— an idea, a phrase, an image—go viral? For starters, the meme must have broad appeal, so it can spread not just within communities of like-minded individuals but can leap from one community to the next. Researchers, by mining public Twitter data, have found that a meme's “virality” is often evident from the start. After only a few dozen tweets, a typical viral meme (as defined by tweets using a given hashtag) will already have caught on in numerous communities of Twitter users. In contrast, a meme destined to peter out will resonate in fewer groups.
“We didn't expect to see that the viral memes were going to behave very differently from nonviral memes at their beginnings,” says Lilian Weng, a graduate student in informatics at Indiana University Bloomington. Those differences allowed Weng and her colleagues to forecast memes that would go viral with an accuracy of better than 60 percent, the team reported in a 2013 study.
Images become an important and prevalent way to express users' activities, opinions and emotions. In a social network, individual emotions may be influenced by others, in particular by close friends. We focus on understanding how users embed emotions into the images they uploaded to the social websites and how social influence plays a role in changing users' emotions. We first verify the existence of emotion influence in the image networks, and then propose a probabilistic factor graph based emotion influence model to answer the questions of "who influences whom". Employing a real network from Flickr as experimental data, we study the effectiveness of factors in the proposed model with in-depth data analysis. Our experiments also show that our model, by incorporating the emotion influence, can significantly improve the accuracy (+5%) for predicting emotions from images. Finally, a case study is used as the anecdotal evidence to further demonstrate the effectiveness of the proposed model.
"Within graph theory and network analysis, centrality of a vertex measures the relative importance of a vertex within a graph. The centrality plays key role in network analysis and has been widely studied using different methods. Inspired by the idea of vertex centrality, a novel centrality guided clustering (CGC) is proposed in this paper. Different from traditional clustering methods which usually choose the initial center of a cluster randomly, the CGC clustering algorithm starts from a “LEADER”—a vertex with the highest centrality score—and a new “member” is added into the same cluster as the “LEADER” when some criterion is satisfied. The CGC algorithm also supports overlapping membership. Experiments on three benchmark social network data sets are presented and the results indicate that the proposed CGC algorithm works well in social network clustering."
In this work, a novel hierarchical clustering algorithm is proposed for social network clustering. Traditional clustering methods, such as -means, usually choose clustering centers randomly, and the hierarchical clustering algorithms usually start from two elements with shortest distance. Different from these methods, this work chooses the vertex with highest centrality score as the starting point. If one does some analysis on social network datasets, one may notice that in each community, there is usually some member (or leader) who plays a key role in that community. In fact, centrality is an important concept  within social network analysis. High centrality scores identify members with the greatest structural importance in a network and these members are expected to play key roles in the network. Based on this observation, this work proposes to start clustering from the member with highest centrality score. That is, a group is formed starting from its “leader,” and a new “member” is added into an existing group based on its total relation with the group. The main procedure is as follows. Choose the vertex with the highest centrality score which is not included in any existing group yet and call this vertex a “LEADER.” A new group is created with this “LEADER.” Repeatedly add one vertex to an existing group if the following criterion is satisfied: the density of the newly extended group is above a given threshold.