CAREER: Machine Learning and Event Detection for the Public Good

本研究的目标是创建和探索在大量复杂的现实世界数据集中检测新事件的新方法。该方法包括新算法，以有效和准确地找到大型高维数据集的最异常子集，以及方法上的进步，以将来自用户反馈的增量模型学习整合到事件检测中，整合来自新兴的社会规模数据，变革技术，如手机和用户生成的网络内容，并通过创建方法和工具来增强事件检测，用于事件表征，解释，可视化，调查和回应。实验研究与多管齐下的教育计划相结合，通过开发机器学习和政策研究与教育的课程和研讨会，将机器学习纳入公共政策课程，并建立一个新的机器学习和政策联合博士课程。该项目的成果将纳入已部署的事件监测系统，并应用于公共卫生、执法和卫生保健领域，从而能够更及时和准确地发现新出现的疾病爆发，预测新出现的暴力犯罪热点，并查明病人护理的异常模式。项目结果，包括出版物、软件和数据集，将通过项目网站（http://www.cs.cmu.edu/~neill/CAREER）发布。

The goal of this research is to create and explore novel methods for detection of emerging events in massive, complex real-world datasets. The approach consists of new algorithms to efficiently and exactly find the most anomalous subsets of a large, high-dimensional dataset, as well as methodological advances to incorporate incremental model learning from user feedback into event detection, incorporate society-scale data from emerging, transformative technologies such as cellular phones and user-generated web content, and augment event detection by creating methods and tools for event characterization, explanation, visualization, investigation and response. The experimental research is integrated with a multi-pronged educational initiative to incorporate machine learning into the public policy curriculum through development of courses and seminars, workshops in machine learning and policy research and education, and establishment of a new Joint Ph.D. Program in Machine Learning and Policy. The results of this project will be incorporated into deployed event surveillance systems and applied to the public health, law enforcement, and health care domains, enabling more timely and accurate detection of emerging outbreaks of disease, prediction of emerging hot-spots of violent crime, and identification of anomalous patterns of patient care. Project results, including publications, software, and datasets, will be disseminated via project web site (http://www.cs.cmu.edu/~neill/CAREER).