David Drew, John Parlee
United States Military Academy
Subject Listing - Computer Science
Advisor: Major Fernando Maymi
Thursday, Oral Session 1, Presentation 5, Owen Hall 202
CLADE: A TEXT ATTRIBUTION SYSTEM FOR LAW ENFORCEMENT
The vast expanse of technology in today's world grants the ability to send and receive electronic or handwritten messages without being able to identify a source of origin. We propose a system that stores, tracks, and analyzes text within these messages. Text is found in different forms, ranging from e-mails to hand written notes. Our system inputs text, stores it, analyzes data and language, and then determines commonalities with other text sources. This project uses a variety of user selected comparison methods to compare one text against a variety of other texts within the stored database. The database is accessed using a World Wide Web interface. Based on user selected models the system returns a result stating the probability that the comparison text was written by the same author as another piece of text in the database. The user can then take this information and make and informed judgment whether on not the percentage is high enough to consider the documents to be from the same author. The results of these comparisons will potentially allow one to discover the origin and author of these text samples. Though our system goals parallel those of existing tools that detect plagiarism, Clade is unique in that it takes into account a variety of other parameters such as the internet protocol addresses of messages and particular idioms that may indicate the regional background of an author. This system was developed in response to a problem expressed by a Federal law enforcement agency in the United States. This agency was looking for novel ways to determine if multiple threatening email messages shared a common author, or authors with a common background. The ability to determine this would allow the agency to focus its investigative efforts.
Advisor: Major Fernando Maymi, Assistant Professor, Department of EE&CS, United States Military Academy, West Point, NY