II-NEW: Infrastructure for Change: From a Teaching Department to National Prominence

Georgetown University, Washington DC

Investigators

Abstract

Over the last several years, Georgetown University has espoused a policy of growth for its science departments. This project supports Georgetown University's growth goals by positing a multifaceted research program that explores new techniques for processing large collections of text data. The project examines how both new and existing text processing methods may be applied to a variety of open problems across different disciplines within computer science. Research activities include text mining for large-scale document collections; machine learning techniques for detecting malicious executables; information retrieval methods for performing proactive digital forensic investigations; text processing approaches for preventing privacy breaches in encrypted voice-over-IP (VoIP) communication; and data mining and sentiment detection methods for automated content analysis of large quantities of political science texts. To support these goals, the project includes the creation of a next-generation infrastructure for large-scale text processing. Data-intensive tasks will use interconnected front-end systems, connected via high-speed networking to scalable back-end storage solutions. The infrastructure supports the development and testing of novel text processing algorithms on very large corpora. The project's themes of text processing and information assurance have been integrated into the undergraduate and graduate curricula at Georgetown University. More broadly, the project benefits society by improving the ability to (1) process large quantities of information, (2) identify malicious applications, and (3) improve the resiliency of digital communication networks to eavesdropping attacks. The project broadens the participation in research through the University's Summer Programs for High School Students and the Center for Multicultural Equity and Access.