ترغب بنشر مسار تعليمي؟ اضغط هنا

On the Convexity of Latent Social Network Inference

185   0   0.0 ( 0 )
 نشر من قبل Seth Myers
 تاريخ النشر 2010
والبحث باللغة English




اسأل ChatGPT حول البحث

In many real-world scenarios, it is nearly impossible to collect explicit social network data. In such cases, whole networks must be inferred from underlying observations. Here, we formulate the problem of inferring latent social networks based on network diffusion or disease propagation data. We consider contagions propagating over the edges of an unobserved social network, where we only observe the times when nodes became infected, but not who infected them. Given such node infection times, we then identify the optimal network that best explains the observed data. We present a maximum likelihood approach based on convex programming with a l1-like penalty term that encourages sparsity. Experiments on real and synthetic data reveal that our method near-perfectly recovers the underlying network structure as well as the parameters of the contagion propagation model. Moreover, our approach scales well as it can infer optimal networks of thousands of nodes in a matter of minutes.



قيم البحث

اقرأ أيضاً

Although social neuroscience is concerned with understanding how the brain interacts with its social environment, prevailing research in the field has primarily considered the human brain in isolation, deprived of its rich social context. Emerging wo rk in social neuroscience that leverages tools from network analysis has begun to pursue this issue, advancing knowledge of how the human brain influences and is influenced by the structures of its social environment. In this paper, we provide an overview of key theory and methods in network analysis (especially for social systems) as an introduction for social neuroscientists who are interested in relating individual cognition to the structures of an individuals social environments. We also highlight some exciting new work as examples of how to productively use these tools to investigate questions of relevance to social neuroscientists. We include tutorials to help with practical implementation of the concepts that we discuss. We conclude by highlighting a broad range of exciting research opportunities for social neuroscientists who are interested in using network analysis to study social systems.
Population behaviours, such as voting and vaccination, depend on social networks. Social networks can differ depending on behaviour type and are typically hidden. However, we do often have large-scale behavioural data, albeit only snapshots taken at one timepoint. We present a method that jointly infers large-scale network structure and a networked model of human behaviour using only snapshot population behavioural data. This exploits the simplicity of a few parameter, geometric socio-demographic network model and a spin based model of behaviour. We illustrate, for the EU Referendum and two London Mayoral elections, how the model offers both prediction and the interpretation of our homophilic inclinations. Beyond offering the extraction of behaviour specific network structure from large-scale behavioural datasets, our approach yields a crude calculus linking inequalities and social preferences to behavioural outcomes. We give examples of potential network sensitive policies: how changes to income inequality, a social temperature and homophilic preferences might have reduced polarisation in a recent election.
How people connect with one another is a fundamental question in the social sciences, and the resulting social networks can have a profound impact on our daily lives. Blau offered a powerful explanation: people connect with one another based on their positions in a social space. Yet a principled measure of social distance, allowing comparison within and between societies, remains elusive. We use the connectivity kernel of conditionally-independent edge models to develop a family of segregation statistics with desirable properties: they offer an intuitive and universal characteristic scale on social space (facilitating comparison across datasets and societies), are applicable to multivariate and mixed node attributes, and capture segregation at the level of individuals, pairs of individuals, and society as a whole. We show that the segregation statistics can induce a metric on Blau space (a space spanned by the attributes of the members of society) and provide maps of two societies. Under a Bayesian paradigm, we infer the parameters of the connectivity kernel from eleven ego-network datasets collected in four surveys in the United Kingdom and United States. The importance of different dimensions of Blau space is similar across time and location, suggesting a macroscopically stable social fabric. Physical separation and age differences have the most significant impact on segregation within friendship networks with implications for intergenerational mixing and isolation in later stages of life.
66 - Dmitry Zinoviev 2020
Kompromat (the Russian word for compromising material) has been efficiently used to harass Russian political and business elites since the days of the USSR. Online crowdsourcing projects such as RuCompromat made it possible to catalog and analyze kom promat using quantitative techniques -- namely, social network analysis. In this paper, we constructed a social network of 11,000 Russian and foreign nationals affected by kompromat in Russia in 1991 -- 2020. The network has an excellent modular structure with 62 dense communities. One community contains prominent American officials, politicians, and entrepreneurs (including President Donald Trump) and appears to concern Russias controversial interference in the 2016 U.S. presidential elections. Various network centrality measures identify seventeen most central kompromat figures, with President Vladimir Putin solidly at the top. We further reveal four types of communities dominated by entrepreneurs, politicians, bankers, and law enforcement officials (siloviks), the latter disjointed from the first three.
We deal with the problem of automatically generating social networks by analyzing and assessing smartphone usage and interaction data. We start by assigning weights to the different types of interactions such as messaging, email, phone calls, chat an d physical proximity. Next, we propose a ranking algorithm which recognizes the pattern of interaction taking into account the changes in the collected data over time. Both algorithms are based on recent findings from social network research.
التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا