TWC: Small: Collaborative: Reputation-Escalation-as-a-Service: Analyses and Defenses

TWC：小型：协作：声誉升级即服务：分析和防御

• 批准号: 1618117
• 负责人: Haining Wang
• 金额: $ 20万
• 依托单位: University of Delaware
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-07-01 至 2020-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1618117&HistoricalAwards=false
• 关键词: TWC; Small; Collaborative; Reputation; Escalation

Living in an age when services are often rated, people are increasingly depending on reputation of sellers or products/apps when making purchases online. This puts pressure on people to gain and maintain a high reputation by offering reliable and high-quality services and/or products, which benefits the society at large. Unfortunately, due to extremely high competition in e-commerce or app stores, recently reputation manipulation related services have quickly developed into a sizable business, which is termed Reputation-Escalation-as-a-Service (REaaS). As REaaS attacks grow in scale, effective countermeasures must be designed to detect and defend against them. This research addresses REaaS from two aspects. First, it aims to understand the economics of REaaS by conducting empirical studies of e-markets. Second, it aims to develop defensive measures, which involve both technical approaches and market intervention. The technical approaches focus on detection of REaaS from e-markets, and novel detection techniques will be developed using content analysis, machine learning, social ties, and graph theory. For market invention, after a holistic analysis of REaaS, this research aims to identify its bottleneck (the weakest link) and also measure the efficacy of intervention. The outcome of this data-driven security research will enhance security education with labs based on social-economic data analysis. The success of this research will attract more attention of industry practitioners, government sectors, and academia to jointly tackle the REaaS problem.

生活在服务经常被评为服务的时代，人们越来越多地取决于卖方或产品/应用在线购买时的声誉。这给人们带来了压力，要求人们通过提供可靠和高质量的服务和/或产品来获得高声誉，从而使整个社会受益。不幸的是，由于在电子商务或应用商店中竞争极高的竞争，最近与声誉操纵相关的服务迅速发展为一项巨大的业务，该业务被称为声誉 - 估计-AS-A-Service（Reaas）。随着Reaas攻击的规模增长，必须设计有效的对策来检测和防御它们。这项研究涉及两个方面的Reaas。首先，它旨在通过对电子市场进行实证研究来了解Reaas的经济学。其次，它旨在制定防御措施，涉及技术方法和市场干预。 技术方法的重点是从电子市场中检测REAA，并且将使用内容分析，机器学习，社交关系和图理论开发新的检测技术。对于市场发明，在对Reaas进行了整体分析之后，该研究旨在确定其瓶颈（最弱的联系），并衡量干预的功效。这项数据驱动的安全研究的结果将通过基于社会经济数据分析的实验室来增强安全教育。这项研究的成功将吸引行业从业人员，政府部门和学术界的更多关注，共同解决Reaas问题。