Collaborative Research: III: Small: Entity- and Event-driven Media Bias Detection

民主是由公众舆论塑造的，而公众舆论又受到人们阅读、观看和收听的新闻的显著影响。因此，对于知情的公众来说，了解他们所消费的新闻是如何被选择、包装和呈现的是至关重要的。该项目旨在建立计算系统，以检测和量化媒体意识形态如何影响文章及其组成事件层面的新闻创作和呈现。该项目将促进新闻生产的透明度，提高公众对媒体决策的认识。所开发的工具可以有效和高效地支持在组织和文章层面对媒体意识形态进行测量，从而促进包括政治学、社会科学和传播学在内的广泛领域的研究。拟议的研究将涉及来自不同背景的研究生和本科生，特别是代表性不足的群体。开发的数据集和方法将构成新开发课程模块的基础。项目中产生的知识将通过演示、发表的博客、播客演讲和报纸的客座文章等方式向公众发布。该项目将研究媒体偏见如何通过新闻文章中呈现的内容的选择和组织来包装新闻，并开发实体和事件驱动的计算模型，用于检测意识形态内容选择和预测文章层面的意识形态。将探讨三个主要的研究任务。首先，将开发话语感知事件分类模型，以区分对主要事件的描述与其他上下文通知事件和间接相关事件的描述。其次，将构建实体和事件驱动的上下文表示学习框架，通过捕获实体和事件之间的关系来检测媒体偏见。第三，将研究对抗性学习，通过将媒体特定语言与意识形态内容分离开来，预测具有细粒度得分的新闻文章的政治意识形态。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Democracy is shaped by public opinion, and public opinion in turn is significantly influenced by the news that is read, watched, and listened to. It is thus essential for an informed public to understand how the news they consume is being selected, packaged, and presented. This project aims to build computational systems to detect and quantify how media ideology affects the creation and presentation of news at the level of articles and their constituent events. This project will promote the transparency of news production and enhance public awareness of media decisions. The developed tools can effectively and efficiently support the measurement of media ideology at organization- and article-levels, which facilitates research in broad areas, including political science, social science, and communications. The proposed research will involve graduate and undergraduate students from a diverse array of backgrounds, especially underrepresented groups. The developed datasets and methods will form the basis of modules in newly developed courses. The knowledge produced in the project will be distributed to the public via demos, published blogs, talks at podcasts, and guest essays to newspapers. This project will examine how media bias can result from the packaging of news via the selection and organization of contents presented in news articles, and develop entity- and event-driven computational models for detecting ideological content selection and predicting article-level ideology. Three main research tasks will be explored. First, discourse-aware event categorization models will be developed to distinguish descriptions of main events from other context-informing events and indirectly-related events. Second, an entity- and event-driven contextual representation learning framework will be built to detect media bias by capturing relations between entities and events. Third, adversarial learning will be investigated to predict the political ideology of a news article with a fine-grained score by disentangling media-specific languages from ideological content.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.