Management Science and Information Systems
Department @ Rutgers University

Jaideep S. Vaidya

Professor of Computer Information Systems

Rutgers, The State University of New Jersey
Management Science and Information Systems Department
1 Washington Park
Newark, New Jersey, 07102-1897
Office: WP 1080
Office Phone: +1 973-353-1441
FAX: +1 973-353-5003

Biographical Information


Curriculum Vitae (also in postscript, and pdf).

Notes to students pursuing a Ph.D., Master's, Independent Study, or desiring admission to Rutgers.

Current Areas of Research

See referenced publications for more details, or C. V. for full publication list.

Privacy-Preserving Data Mining

Data is ubiquitous, but knowledge is scarce. Data mining - finding interesting patterns from data, relies on the collection of massive amounts of data, often from multiple different sites. However, privacy/security issues often restrict access to data. Is it possible to mine data when access to it is restricted? My goal is to develop technology to address this in the distributed case: such that only data owners or their representatives have true access to data, but accurate mining results are still computed. We develop algorithms that share some information to calculate correct results, where we can show that the shared information does not disclose private data.

Selected Presentations and Publications:

Privacy Preserving K-Means Clustering on Vertically Partitioned Data, invited talk at Interface '04, the best of data mining at KDD session, Baltimore, Maryland, May 27, 2004.

Privacy Preserving Data Mining over Vertically Partitioned Data at the CSIS Seminar at the Department of Information and Software Engineering, George Mason University, May 25, 2004.

Privacy Preserving Data Mining over Vertically Partitioned Data at the DAIS Seminar at the CS Department at University of Illinois at Urbana-Champaign, October 24, 2003.

Secure Set Intersection Cardinality with Application to Association Rule Mining. Jaideep Vaidya and Chris Clifton, accepted for publication in Journal of Computer Security, IOS Press.

Privacy-Preserving Data Mining: Why, How, and What For?. Jaideep Vaidya and Chris Clifton, accepted for publication in IEEE Security & Privacy, New York, NY.

Privacy Preserving Outlier Detection, Jaideep Vaidya and Chris Clifton, in the 2004 IEEE International Conference on Data Mining, November 2004, Brighton, UK.

Privacy-Preserving K-Means Clustering over Vertically Partitioned Data. Jaideep Vaidya and Chris Clifton, The Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 24 - 27, 2003, Washington, D.C. Honorable mention (runner up), best research paper.
(Revised paper invited for submission at Interface '04).


List of resources in this field courtesy Stanley Oliveira. Stanford PORTIA project reading list.

Other Data Mining and Security Topics

Privacy issues in distributed data mining is only one area where data mining and security interact. Other areas of research include security concerns posed by data mining results (the data isn't private, but what might be learned from it is) and applications of data mining to security (e.g., intrusion detection).

Spatio-Temporal Data Mining

Spatio-Temporal Data Mining raises a whole new set of problems. Spatial data is complex in character, and requires new techniques for doing data mining.

Students and Collaborators

Postdoctoral Opportunities

None funded at this time.

Potential Independent Study Projects

If you would like to pursue a independent study with me, please see the instructions on proposing such study. Occasionally I have specific projects that may be of interest, often involving collaboration with corporate partners. These are listed below: