Scalable Data Exploration and Novelty Detection
journal contributionposted on 2004-01-01, 00:00 authored by Jaime G. Carbonell, Eugene Fink, Chun Jin, Bora Cenk Gazen, J. Mathew, A. Saxena, S Ananthraman, D. Dietrich, G Mani, J. Tittle, P. Durbin
Project ARGUS is focused on helping an analyst explore massive, structured data. This scalable exploration includes exact and partial match queries, monitoring hypotheses and discovery of novel patterns in both static and streaming data. We provide these facilities within the context of an analyst workbench interface called Data Explorer. The remainder of this paper comprises three sections; a) a brief review of the methodology employed within ARGUS for the detection of novelty within massive data, b) monitoring of streaming data and a synopsis of the research originating out of the CMU team on the incremental aggregation on multiple continuous queries, and, c) the development of an analyst workbench environment, called ARGUS Data Explorer that has been developed by the teams from DYNAMiX and ManTech CSEC. The ARGUS Data Explorer is currently being evaluated through the RDEC environment/process as a precursor to possible deployment into live operating environments.