News

June 2008: Paper titled "Troubleshooting Thousands of Jobs on Production Grids Using Data Mining Techniques" accepted at the 9th IEEE/ACM International Conference on Grid Computing (GRID '08). The paper is authored by David Cieslak, Nitesh Chawla, and Douglas Thain.

June 2008: Paper titled "Learning Decision Trees for Unbalanced Data" accepted at the 18th European Conference on Machine Learning (ECML '08) in Antwerp, Belgium. The paper is authored by David Cieslak and Nitesh Chawla.

May 2008: David Cieslak presents research on class imbalance problems at Sandia National Laboratories in Livermore, CA.

More News

About

We apologize for not all information of all our projects/publications available on the website. We will be updating it shortly. In the meantime, please email us at dial@nd.edu with any questions regarding our projects/papers.

Welcome to DIAL (Data, Inference Analysis, and Learning) -- Professor Nitesh Chawla's research group.

"Print, film, magnetic, and optical storage media produced about 5 exabytes of new information in 2002. Ninety-two percent of the new information was stored on magnetic media, mostly in hard disks. " ~Berkeley 2003

This data pervades every sector of human interest, medicine, biology, commerce, WWW, security, fraud, space research, and others, and within it are (potentially) useful information, patterns, associations, and even anomalies. Our main research focus is on the process of automatic discovery and synthesizing useful knowledge from large sets of data. This work taps into many frameworks from related fields such as data mining, machine learning, pattern recognition, databases, probability, statistics, information theory and visualization. Such a broad framework allows the application into a variety of real-world applications.

Our work ranges a diverse set of problems including class imbalance, data segmentation, bias detection, parallel data mining, graph mining and clustering, collaborative filtering, and biometrics.

page hit 
counter