Information Services and Process Innovation Lab

Our research program in data mining and machine learning has been active for over a decade. The team has made interesting and valuable contributions in categorization, quantification, clustering, and feature selection, to name a few. We work with HP's business units - for example, with HP Services in mining call-center records and in managing, categorizing and migrating millions of support documents; and with HP StorageWorks in using categorization to enhance storage systems. We also work with a variety of leading-edge external customers and several universities on exciting new research opportunities.

Selected Publications:

• Pragmatic Text Mining: Minimizing Human Effort to Quantify Many Issues in Call Logs G. Forman, E. Kirshenbaum, J. Suermondt. KDD'06. HPL-2006-60R1.

• Quantifying Trends Accurately Despite Classifier Error and Class Imbalance. G. Forman. KDD'06. HPL-2006-48R1.

• Tackling Concept Drift by Temporal Inductive Transfer. G. Forman. SIGIR'06. HPL-2006-20R1.

• Counting Positives Accurately Despite Inaccurate Classification. G. Forman. ECML'05. HPL-2005-96R1.

• Beware the Null Hypothesis: Critical Value Tables for Evaluating Classifiers. G. Forman & Ira Cohen. ECML'05. HPL-2005-70.

• Learning from Little: Comparison of Classifiers Given Little Training. G. Forman & Ira Cohen. ECML'04. HPL-2004-19R1.

• A Pitfall and Solution in Multi-Class Feature Selection for Text Classification. G. Forman. ICML'04. HPL-2004-86.

• An Extensive Empirical Study of Feature Selection Metrics for Text Classification. G. Forman. Special Issue on Variable and Feature Selection, Journal of Machine Learning Research, 3(Mar):1289-1305, 2003. HPL-2002-147R1.

• Choose Your Words Carefully: An Empirical Study of Feature Selection Metrics for Text Classification. G. Forman. In the Joint Proceedings of the 13th European Conference on Machine Learning and the 6th European Conference on Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD '02), August 19-23, 2002. HPL-2002-88R2.