SaTC: CORE: Small: Interdisciplinary Models to Identify and Understand Cyberbullying

网络欺凌是一种常见的网络不良行为，会造成严重的心理和身体伤害。这个项目整合了计算机科学和心理学框架，以促进我们对如何识别和预防网络欺凌的理解。这个跨学科项目的核心目标是设计和开发模型和应用程序，以防止和识别社交网络中的网络欺凌事件。它使用心理学理论来指导对网络欺凌性质的调查，并采用自动反欺凌工具，以及精心设计的数据收集程序和评估框架。该项目的影响包括共享研究和教育资源，提高政策制定者对网络欺凌的认识，并为研究生和本科生提供科学支撑，以发展成为公认的跨学科学者。这个项目的智力价值来自于它将计算机和心理科学协同整合起来，以解决一个重大的社会问题。该项目研究利用大量心理学相关经验工作的网络欺凌检测的自动化模型，并寻求开发基于证据的工具来识别和预防网络欺凌。它将用于识别网络欺凌的创新机器学习模型(包括个性化、基于时间分析的多模式和隐私保护模型)与用于识别和解决网络欺凌的基于证据的应用程序(例如，在网络欺凌和社交媒体背景下加强亲子关系的应用程序)相结合。该项目还使用系统注释的数据集和调查方法来确定提高反网络欺凌技术的使用和可用性的机制，并调查网络欺凌性质的关键方面，包括网络欺凌的时间特征以及社交媒体使用和网络欺凌的心理相关性之间的联系。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Cyberbullying, a common form of online misbehavior, can do serious psychological and physical harm. This project integrates computer science and psychology frameworks to advance our understanding of how cyberbullying can be identified and prevented. The core objective of this interdisciplinary project is to design and develop models and apps to prevent and identify instances of cyberbullying in social networks. It uses psychological theory to guide investigations of the nature of cyberbullying and the adoption of automated anti-bullying tools, as well as carefully-designed data collection processes and evaluation frameworks. The project's impact includes sharing research and educational resources, raising cyberbullying awareness with policy-makers, and providing graduate and undergraduate students with the scientific scaffolding to develop into recognized interdisciplinary scholars. The intellectual merit of this project stems from its synergistic integration of computer and psychological sciences to address a major social problem. This project studies automated models for cyberbullying detection that leverage a large body of relevant empirical work in psychology and seeks to develop evidence-based tools for identifying and preventing cyberbullying. It integrates innovative machine learning models for cyberbullying identification (including personalized, temporal-analysis-based, multi-modal, and privacy-preserving models) with the development of evidence-based apps for identifying and addressing cyberbullying (e.g., apps that strengthen parent-teen relationships within the context of cyberbullying and social media). The project also uses systematically-annotated datasets and survey methods to identify mechanisms to enhance the use and usability of anti-cyberbullying technologies and investigate key aspects of the nature of cyberbullying, including temporal characteristics of cyberbullying and connections between social media use and psychological correlates of cyberbullying.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.