Current projects
Title: Anomaly Detection for Traffic Monitoring Data
PIs: Aryya Gangopadhyay, Vandana Janeja, Andrew Sears
Funding agency: MDoT--State Highway Administration
Duration: 02/24/2009-01/31/2010
Open RA positions: We are currently funding one graduate student for this project. Future positions may be available.
Abstract
Our overall objective in this project is to find anomalous patterns in terms of the amount of invoice from the contractors. Many factors impact the invoice dollars including the type of counts, the geospatial locations of stations allocated, the year in which the jobs are performed etc. In order to find anomalies, we propose to analyze the data in several ways, which should incrementally enable us to identify the anomalies and the potential factors behind them. We plan on performing our analysis in three stages. Each stage will allow us to gain more insights into the data and provide input to model and perform the tasks in the subsequent stages. In the first stage, we have developed a multidimensional model and by using online analytical processing (OLAP) with operators such as slice, dice, and drill-down we try to identify anomalous data points that clearly stand out from the rest. In the second stage we will apply data mining techniques to analyze the data further and identify clusters as well as anomalies. In the third stage we will apply spatial data mining to take into consideration the spatial distribution of the stations in addition to the other factors.
Title: IIS-IPS: A Privacy-Preserving Framework for Distance-Based Mining
Proposal Number: IIS-IPS-0713345
PIs: Zhiyuan Chen, Aryya Gangopadhyay
Funding agency: NSF
Duration: 09/01/2007-08/31/2010
Open RA positions: We have RA positions for two PhD students and one undergraduate student for this project.
Abstract
In this project we develop a novel framework for preserving privacy in distance based data mining. This research helps to protect the privacy of sensitive data and at the same time allow the discovery of interesting patterns using distance based mining algorithms. The approach in this project consists of a pre-processing step to de-correlate the data and an additive perturbation step to provide worst-case privacy guarantees. This approach also provides the necessary and sufficient conditions for such guarantees. In addition, modifications are made to existing distance-based data mining algorithms so that these algorithms can run accurately on the perturbed data. The results of this project provide privacy preserving data mining techniques with both worst-case privacy guarantees and high accuracy of mining results. These techniques have possible applications in many areas, especially in critical areas such as law enforcement where finger prints, foot prints, and facial images are matched using distance-based algorithms. This research is also linked to educational goals through dissemination of the results to K-12 educational and outreach programs, undergraduate and graduate courses, and interdisciplinary conferences and workshops. The results of this project will be disseminated via the project website Privacy Reseach.
---------------------------------------------------------------------------------------------------------------------------
Title: Graduate Assistance in Areas of National Needs (GAANN)
PIs: Aryya Gangopadhyay, Ant Ozok, Andrew Sears, Donsong Zhang, Lina Zhou
Funding agency: US Department of Education
Duration: 09/01/2004-08/31/2009
Open RA positions: We have several fellowsips positions available for this project.
Abstract
This is a training grant designed to support PhD students in Information Systems through fellowships that include competitive stipend, tuition, health benefits, and support for computer equipement and travel. GAANN fellows must be either US citizens or permanent residents and demonstrate financial eligibility.
---------------------------------------------------------------------------------------------------------------------------
Title: IGERT: Water in the Urban Environment
My role : Affiliated Faculty
Funding agency: NSF
Duration: 09/01/2006-08/31/2011
Home page: IGERT at UMBC
==================================================================================
Past projects
Title: Maryland Voter Information Clearinghouse
PIs: Don Norris, Aryya Gangopadhyay, Andrew Sears
Funding agency: Maryland State Board of Elections
Duration: 2006-2007
Home page: In this project we developed a comprehensive searchable Web-based system for information about polling places and sample ballots, campaign finance, and candidates for elected offices in the State of Maryland. This project funded two PhD and two MS students. The project was subsequently moved to Maryland Voter Information Clearinghouses in 2008.
---------------------------------------------------------------------------------------------------------------------------
Title: Spatiotemporal Data Explorer: An Integrated Spatiotemporal Data Warehouse for Visualization Driven Knowledge Discovery in Water Resources Management
PIs: Michael McGuire, Aryya Gangopadhyay, Zhiyuan Chen, George Karabatis
Funding agency: USGS and National Park Services
Duration: 2006-2007
Abstract
There were two major goals in this project. The first was to create an integrated spatiotemporal data warehouse of water quality and biological resources data created for Maryland but scalable to the area covered by the NBII MAIN. The second was an Internet-based decision support tool for visualizing and mining the integrated data warehouse across spatial and temporal dimensions.
----------------------------------------------------------------------------------------------------------------------------
Title: Content Management User Interface Development
PIs: Aryya Gangopadhyay, Carolyn Seaman (Phase II); Aryya Gangopadhyay (Phase I)
Funding agency: Maryland Industrial Partnerships
Duration: 2002-2003 (Phase II), 2001-2002 (Phase I)
Abstract
This research was a content-based data conversion project funded by the Maryland Industrial Partnerships (MIPS) program from 2001-2003. The project was done in conjunction with Datastreams Inc. in College Park. The project funded one PhD and two Masters students. The project dealt with developing a methodology for converting unstructured text documents into XML documents with user-specified DTD.
----------------------------------------------------------------------------------------------------------------------------
Title: A Web-Based system for Simple Modeling of Nonpoint pollution in the Hackensack Meadowland District
PIs: Aryya Gangopadhyay
Funding agency: Meadowlands Environmental Research Institute
Duration: 2000
Abstract
The objective of this project was to develop a prototype of Internet based Spatial Decision Support System (SDSS) software to support users at different locations and different time to assess the hydrological impact of land use changes. It used an explorative approach to evaluate the hydrological scenarios in response to the land use changes allowing planners and stakeholders to dynamically visualize these impacts. This project funded one PhD student.
----------------------------------------------------------------------------------------------------------------------------