RAPID: SaTC: CORE: Monitoring Social Media for Devising Improved Safeguards Online

在线社交网络促进互动、协作和贡献。相互联系、相互沟通的个体之间的互动促进了集体智慧，特别是当人们致力于一个共同的目标或共同的实践时。最近，由于在线攻击的大幅增加，英语和非英语社交媒体帖子都大幅增加，人们在其中讨论使用各种保护措施来帮助保护他们的在线安全，隐私和安全。该项目正在开发一个框架和方法，用于捕获利用社交媒体平台的个人群体之间共享的新兴集体智慧，以及寻求使用和遵循保障措施以改善其在线安全，隐私和安全的个人群体。 研究这些讨论将增加我们对人们的安全和隐私需求的理解，发现现有保障措施的实际挑战和局限性，并阐明在线安全和隐私研究的未来方向。 该项目将开发一个社会监测框架和方法，用于检测，收集和分类英语和非英语社交媒体帖子和对话，这些帖子和对话围绕新兴的安全漏洞，隐私威胁以及保护人们安全，隐私和安全的保障措施的讨论。该项目还将推进现有的信息检索和自然语言处理方法，包括基于深度学习的分类器、情感、立场和摘要，以便从英语和非英语帖子中提取知识。从这项研究中获得的数据和信息将对试图设计有效，可用，安全和隐私保护方法和工具的研究人员，开发额外保护功能的公司和服务提供商，该奖项反映了NSF的法定使命，并通过使用基金会的智力价值进行评估，被认为值得支持和更广泛的影响审查标准。

Online social networking fosters interaction, collaboration, and contribution. The interactions among interconnected, communicating individuals facilitate collective intelligence, especially when people are engaged in a common goal or a shared practice. Recently, due to the tremendous increase in online attacks, there has been a substantial increase in both English and non-English social media posts in which people discuss the use of various safeguards to help protect their online security, privacy, and safety. This project is developing a framework and approach for capturing the emerging collective intelligence shared among groups of individuals who utilize social media platforms and who seek to use and follow safeguards to improve their online security, privacy, and safety. Studying these discussions will increase our understanding of people’s security and privacy needs, detect the practical challenges and limitations of available safeguards, and shed light on the future direction of online security and privacy research. This project will develop a social monitoring framework and approach for detecting, collecting, and categorizing English and non-English social media posts and conversations around emerging security vulnerabilities, privacy threats, and discussions of safeguards for protecting people's security, privacy, and safety. The project will also advance existing information retrieval and natural language processing methods, including deep learning-based classifiers, sentiment, stance, and summarization in order to extract knowledge from English and non-English posts. The data and information obtained from this study will be highly valuable to researchers trying to design effective, usable, secure, and privacy-preserving methods and tools, companies and service providers developing additional safeguard features, and Internet users seeking to protect their online security and privacy.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.