EAGER: Modeling and Visualization of Latent Communities

EAGER：潜在社区的建模和可视化

• 批准号: 1059577
• 负责人: Peter Brusilovsky
• 金额: $ 15.59万
• 依托单位: University of Pittsburgh
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2010
• 资助国家: 美国
• 起止时间: 2010-09-15 至 2011-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1059577&HistoricalAwards=false
• 关键词: EAGER; Modeling; Visualization; Latent; Communities

In many areas of human professional and social life, people tend to form more or less clearly defined communities. The main problem of these hidden or latent communities is that they are really hard to discover since the borders of these communities cut through various professional and organizational borders. The modern social Web, however, provides a huge volume of alternative data sources for discovering latent communities. The goal of this proposal is to explore a range of promising approaches that can be used to elicit latent communities from various kinds of data about individuals available in the modern social Web and deliver the results for human thinking and interactive exploration through interactive visualizations. The visualization provided will allow humans explore and manipulate the results delivered by the new algorithms. This will deliver results that are produced by the joint power of human and artificial intelligence. In the course of the project, the team will build several data sets combining data of several social Web systems and use these data sets to develop, evaluate, and compare several elicitation and visualization approaches. The work will advance the research on latent communities, community and user modeling, and interactive social visualization. At the same time, the work will constitute one of the first attempts to use a variety of social Web data and a variety of approaches for community modeling. To increase the broader impact of the project, the researcher will apply the latent community knowledge to several practical tasks, such as identifying proper academic mentors and forming coherent collaboration groups. They will also engage a number of students in the research advancing their training into this emerging field.

在人类职业和社会生活的许多领域，人们倾向于形成或多或少明确定义的社区。这些隐藏或潜在社区的主要问题是，它们很难被发现，因为这些社区的边界跨越了各种专业和组织的边界。然而，现代社交网络为发现潜在社区提供了大量的可选数据源。本提案的目标是探索一系列有前途的方法，这些方法可用于从现代社交网络中可用的关于个人的各种数据中引出潜在的社区，并通过交互式可视化将结果交付给人类思考和交互式探索。提供的可视化将允许人类探索和操作新算法提供的结果。这将产生由人类和人工智能共同力量产生的结果。在项目过程中，团队将结合几个社会网络系统的数据构建几个数据集，并使用这些数据集来开发、评估和比较几种启发和可视化方法。这项工作将推动潜在社区、社区和用户建模以及交互式社会可视化的研究。同时，这项工作将构成使用各种社交网络数据和各种社区建模方法的首次尝试之一。为了增加项目的广泛影响，研究人员将把潜在的社区知识应用于几个实际任务，例如确定适当的学术导师和形成连贯的协作小组。他们还将让一些学生参与研究，促进他们在这一新兴领域的培训。