This award supports participants in a two-day symposium on machine learning for anomaly detection to be held at Stanford University on May 22-23, 2004. The purpose of the symposium is to bring together researchers, students, and industrial practitioners from different application areas and let them report their recent results and discuss common concerns. A central aspect of the symposium are eleven invited talks given by established researchers whose work provides a good sampling of research using machine learning methods to detect anomalies across a variety of domains. The symposium fosters improved research in anomaly detection by exposing attendees to ideas from other fields, encouraging interaction and collaboration, and helping to identify common open problems. Dissemination plans include developing a website (http://cll.stanford.edu/symposia/anomaly/) with the speakers talks and relevant papers available for download, preparing a workshop report, and organizing a special journal issue devoted to machine learning approaches to anomaly detection. Finally, the symposium will contribute to increased understanding of methods for anomaly detection, which is becoming important in many areas such as public health monitoring, computer network security, and accounting fraud detection.
