International audience ; We study a stochastic model of influence where agents have "yes" or "no" inclinations on some issue, and opinions may change due to mutual influence among the agents. Each agent independently aggregates the opinions of the other agents and possibly herself. We study influence processes modeled by ordered weighted averaging operators, which are anonymous: they only depend on how many agents share an opinion. For instance, this allows to study situations where the influence process is based on majorities, which are not covered by the classical approach of weighted averaging aggregation. We find a necessary and sufficient condition for convergence to consensus and characterize outcomes where the society ends up polarized. Our results can also be used to understand more general situations, where ordered weighted averages are only used to some extent. Furthermore, we apply our results to fuzzy linguistic quantifiers, i.e., expressions like "most" or "at least a few".
International audience ; We study a stochastic model of influence where agents have "yes" or "no" inclinations on some issue, and opinions may change due to mutual influence among the agents. Each agent independently aggregates the opinions of the other agents and possibly herself. We study influence processes modeled by ordered weighted averaging operators, which are anonymous: they only depend on how many agents share an opinion. For instance, this allows to study situations where the influence process is based on majorities, which are not covered by the classical approach of weighted averaging aggregation. We find a necessary and sufficient condition for convergence to consensus and characterize outcomes where the society ends up polarized. Our results can also be used to understand more general situations, where ordered weighted averages are only used to some extent. Furthermore, we apply our results to fuzzy linguistic quantifiers, i.e., expressions like "most" or "at least a few".
This report describes Iran's influence over the post-Saddam government in Iraq. A key concern is that Iran is extending support for related militias in Iraq that are now responsible for much of the sectarian violence in Iraq.
In this manuscript, we introduce a new measure of political influence in California. Leveraging a new dataset of candidate rankings of their own endorsements, we use the Bradley-Terry model to estimate influence for a broad array of officeholders, interest groups, and endorsing organizations who participate in California politics. We call this new measure of influence Clout scores. Our measure of a person's clout is based on how much candidates for office desire that individual's endorsement. Specifically, we measure a political actor's clout by estimating the extent to which that actor's endorsement is preferred to a baseline endorsement group. Our estimates provide an original, empirically-grounded portrait of the distribution of political capital in California and highlight which political elites have the greatest capacity to swing election outcomes.
Citation: Drollinger, Della. Influence of expositions. Senior thesis, Kansas State Agricultural College, 1902. ; Morse Department of Special Collections ; Introduction: It would not be easy to overestimate the educational value of the objects of beauty, monumental architecture, galleries of painting and sculpture, and museums that are found in some of the old European towns, for such things, stimulate the imagination, increase a thirst for knowledge, and awaken and educate the aesthetic faculties. Much of the marvelous beauty and wealth of public architecture and art in small European towns has been due to earlier political conditions. The town at some time probably was the seat of government of some small kingdom, but in these days of great empires those old time motives have passed away and something new must take their place. The tendency to create expositions is a valuable part of the outworkings of these new motives.
Like any form of human interaction and communication it is possible to view Social Media as a means for the powerful to influence and control the less powerful. But what is power on social media, how might we measure or affect it, and does it translate to the real world? In this lecture we will look at the philosophical definitions of power, and explore how it has been analysed in social networks and social media systems. We will also look at the characteristics of social networks that impact on power, including Homophily, Heterophily, CyberBalkanization and Thresholds of Collective Action. Finally we will ask what evidence there is that power in social media can affect what goes on in the real world, and explore some real and fictional examples of protest to see what the consequences of social media actually are on sometimes violent political debate. Concluding that the power of social media often lies with those operating the network, or with access to the data, rather than the individuals using the system.
This thesis is composed by three standalone papers. The first chapter is about opinion formation processes. Individuals influence each other according to the network structure. If the network is connected and satisfies other mild assumptions, the society will reach a consensus. Therefore, it is a matter of interest understanding when the network would be connected or not. Here, we develop a model where the network takes place endogenously, and agents update their opinions accordingly. We study general conditions on the initial distribution of opinions such that consensus will be reached. We provide sufficient conditions for connectedness. In the dynamic model we show that polarization both in the transition and the long run. This essay is a joint work with Paolo Pin (Bocconi University). The second chapter is about peer effects. I consider how social pressure affects the strategic network formation process. Agents choose their links subject to mutual consent, and then equilibrium behaviors are determined by an underlying game where agents choose their effort. I provide a characterization result for all pairwise Nash stable network, and use farsightedness to avoid cycles. As an application, I consider peer effects in the classroom. Results suggest that we could set optimal policies to improve outcome of low achievement students through targeted incentives schemes. In the last chapter I focus on the interaction between real and virtual life. To do so I propose a model of search where agents choose in equilibrium how intense is their on-line activity. The outcomes of the game depend on the network, which takes place through a semi-random process. I extend the standard random network formation allowing agents to hold a certain degree of choice. The model allows to track characteristics on the meeting environment and individual preferences that would make virtual and real life substitutes or complements.
International audience ; The objective of this paper is to analyze whether the European Union has a strategic vision that would mean a capacity to influence. The reflection is illuminated by the context of migration due to outflows of Mesopotamia started in 2011 and the geopolitical realities. ; L'objectif de ce texte est d'analyser si l'Union européenne a une vision stratégique qui se traduirait par une capacité d'influence. La réflexion est éclairée par le contexte des migrations dues aux exodes de Mésopotamie commencées en 2011 et des réalités géopolitiques.
International audience ; The objective of this paper is to analyze whether the European Union has a strategic vision that would mean a capacity to influence. The reflection is illuminated by the context of migration due to outflows of Mesopotamia started in 2011 and the geopolitical realities. ; L'objectif de ce texte est d'analyser si l'Union européenne a une vision stratégique qui se traduirait par une capacité d'influence. La réflexion est éclairée par le contexte des migrations dues aux exodes de Mésopotamie commencées en 2011 et des réalités géopolitiques.
This thesis is a collection of three stories about influence. Each of them tells, in different ways, how individuals and organizations can be swayed by external agents and what consequences such influence begets.
Each of us has secrets of our own and we know others' secrets too. We share these secrets with some people and we keep these secrets from other people. This affects what we know about each other and how, in turn, we are influenced by each other. Social science scholars have consistently found that people influence each other with regard to matters that can be observed like dropping out of school, weight gain or family structures. But of course, there are whole swaths of social life that are unobservable. The central question of this dissertation is: how do we understand social influence when people keep secrets and share them selectively?Existing formal theories of social influence within social networks examine the structure of the network and the relationships between people. These networks are merely the potential for interpersonal communication and influence. Some work, in particular the work on diffusion, examines not just networks but communication networks. But as this inquiry will reveal, even the communication network is a potential network for the transmission of information with regard to a given topic. This inquiry focuses not on potential communication but on realized communication and how that might differ across discussion partners and topics; it does so because communication is necessary for the influence process to occur. I explore secrets and social influence using the test case of abortion secrets in the United States. Abortion is a highly volatile, contested secret. Tens of millions of American women have had abortions and the legality of abortion sits at the center of American politics. I begin by examining who has abortion secrets, or abortion incidence. I calculate the first set of lifetime abortion incidence measures for birth cohorts in the United States and discuss the implications of cohort rates on public opinion and behavior. Theories of social influence suggest that the tens of millions of women who have had abortions in the United States are themselves social and political actors who can influence others, particularly their peers. Hence, scholarship on abortion as a social and political phenomenon should include incidence rates, particularly cohort incidence rates. I find that the number of lifetime abortions a birth cohort of women has differs by when they were born, even when all the cohorts have spent their entire childbearing years with federally legal abortion. Further, when cohort abortion rates decline, all racial and ethnic groups' cohort rates decline at approximately the same rate. This work is based on vital statistic rates.I move on to how secrets spread and the implications of hearing - or not - others' secrets. I compare abortion and miscarriage secrets; this is analytically helpful for two reasons. First, they are concealable, since both are pregnancy-ending events that occur primarily in the first trimester. Abortion, however, is much more highly stigmatized than miscarriage. Second, having had at least one miscarriage is a near-random occurrence; I can thus exploit it to control for a number of otherwise unobservable characteristics that may affect whether someone hears an abortion secret, such as the prevalence of pregnancy in a respondents' social network and the frequency of discussing those pregnancies.Though abortion is more common and affects more women than miscarriage, many more Americans report knowing someone who has had a miscarriage than an abortion. Furthermore, Americans who are anti-abortion are much less likely than their pro-choice peers to hear abortion secrets and as such think they do not know any woman who has had one. This is likely not the case. Rather, women who have had abortions and the people who know about the abortion elect to keep the abortion secret from people who may disapprove. As such, pro-choice Americans hear abortion secrets and perceive - and hence experience - a more diverse network than people who are anti-abortion and do not hear the secrets. I show that individuals' attitudes determine what kind of community they experience; usually they experience one that aligns with their attitudes. This is not because they have chosen to be with people who agree with them or behave as they like. Rather, it is because the people they are with hide information that would reveal difference. With respect to attributes that can be kept secret, diversity is then not just a characteristic of a community but also a characteristic of individual experience of a community. When individuals keep secrets from those who will disapprove of them, processes of social influence - on public opinion, on tolerance and on behavior - are thwarted.These results are based on a nationally representative survey I designed and conducted of over 1600 American adults who provided information on their experience with and knowledge of others' abortions and miscarriages. They also gave detailed information on disclosing and withholding their own and others' secrets. Using the survey data, I then consider how hearing - or not - women's abortion secrets affects Americans' understanding of who gets abortions and why. We form our understandings of the world, what is possible and what is not, what is appropriate and what is not, on a number of things. We heavily weight our own experiences and those of the people we know. If our information about the experiences of those we know is distorted then that will affect our perceptions of the world, or in this case, of abortion patients. Americans' public opinion regarding legal abortion is highly contingent on who the patient is and her reasons for seeking an abortion. Yet, I show Americans misperceive abortion prevalence, the demographic characteristics of abortion patients and their motivations. Further, their perceptions vary most consistently by the number of abortion secrets they have heard. This suggests Americans infer to the national population of abortion patients based on the patients they know of. But, as shown earlier, whose abortions they know of, is highly dependent on their attitudes. This dissertation documents how abortions are often kept secret, particularly from Americans who are opposed to legalized abortion. As such, Americans misperceive the number of women they know who have had abortions. They then infer from their social network to the nation as a whole and similarly misperceive the abortion patient population nationally. The implications of these misperceptions are wide-ranging; this work focuses on the implications for public opinion.In the final chapter of this dissertation, I outline where I will next take the study of secrets. I will examine a wide variety of secrets, from political attitudes to cancer diagnoses and I will employ both quantitative and qualitative methods. Americans keep abortions secret to avert stigma but there are other reasons why one might keep a secret. By extending into other arenas, I will explore these other reasons. I will also be able to capture how secret-keeping and selective disclosure affect other components of American life including the funding of scientific research and the perceptions of polarization. I hope this dissertation will stand as the first steps toward a comprehensive sociology of secrets.
Dyadic representation has received considerable attention in the US, but much less attention in parliamentary systems where party discipline strongly limits representatives' capacity for individual action. A link between the legislative behaviour of representatives and the preferences of their geographic constituencies may nevertheless exist outside the US, however, particularly in single member plurality systems where the electoral connection is strong. This paper tests for evidence of this dyadic relationship in Question Period in the Canadian Parliament, across three policy domains: defense, debt and taxes, and welfare. As anticipated, there is evidence of dyadic representation in Canada. Results are discussed as they pertain to the comparative study of legislative institutions and political representation.
The advent of digital information technology heralded the concept of information warfare. This 'preliminary' stage in the 1990s really consisted of technology warfare where the networks, upon which combat relied, were seen as weapons to gain 'information superiority'. This was the inception of the technological aspect of Information Warfare. The realisation of the effectiveness of electronic networks to optimize organisational communication was taken up by industry, the military and terrorist groups alike. As society quickly became more reliant on digital networks to run its critical functions, it became apparent that this infrastructure was vulnerable and needed protection (as well as being a target for offensive operations).The next stage was the emphasis not on the technology but the information which it stored and processed. This was the 'information' stage of Information Warfare (now renamed Information Operations to reflect its expanded scope). This stage developed further and gradually started to include elements such as public relations, and strategic communications This paper examines the development of these elements and their use by terrorist groups. It concentrates on the contemporary manifestations of Influence Wars.
International audience ; The problem of maximizing or minimizing the spreading in a social network has become more timely than ever with the advent of fake news and the coronavirus epidemic. The solution to this problem pertains to influence maximization algorithms that identify the right nodes to lockdown for epidemic containment, hire for viral marketing campaigns, block for online political propaganda etc. Though these algorithms have been developed for many years, the majority of the literature focuses on scalability issues and relaxing the method's assumptions. In the recent years, the emergence of new complementary data and more advanced machine learning methods for decision have guided part of the literature towards learning-based approaches. These can range from learning how information spreads over a network, to learning how to solve the combinatorial optimization problem itself. In this tutorial, we aim to dissentangle and clearly define the different tasks around learning for influence applications in social networks. More specifically, we start from traditional influence maximization algorithms, describe the need of influence estimation and delineate the state-of-the-art on influence and diffusion learning. Subsequently, we delve into the problem of learning while optimizing the influence spreading which is based on online learning algorithms. Finally, we describe the latest approaches on learning influence maximization with graph neural networks and deep reinforcement learning.
International audience ; The problem of maximizing or minimizing the spreading in a social network has become more timely than ever with the advent of fake news and the coronavirus epidemic. The solution to this problem pertains to influence maximization algorithms that identify the right nodes to lockdown for epidemic containment, hire for viral marketing campaigns, block for online political propaganda etc. Though these algorithms have been developed for many years, the majority of the literature focuses on scalability issues and relaxing the method's assumptions. In the recent years, the emergence of new complementary data and more advanced machine learning methods for decision have guided part of the literature towards learning-based approaches. These can range from learning how information spreads over a network, to learning how to solve the combinatorial optimization problem itself. In this tutorial, we aim to dissentangle and clearly define the different tasks around learning for influence applications in social networks. More specifically, we start from traditional influence maximization algorithms, describe the need of influence estimation and delineate the state-of-the-art on influence and diffusion learning. Subsequently, we delve into the problem of learning while optimizing the influence spreading which is based on online learning algorithms. Finally, we describe the latest approaches on learning influence maximization with graph neural networks and deep reinforcement learning.
