I'm currently interested in machine learning problems in classification and ranking for the purpose of information filtering. Some of these topics include learning in the presences of noisy or adversarial labels, learning at scale over high dimensional data, and learning small human-interpretable models. For more details, see my home page.
Going Mini: Extreme Lightweight Spam Filters, D. Sculley, Gordon V. Cormack, CEAS 2009: Proceedings of the Sixth Conference on Email and Anti-Spam.
Predicting Bounce Rates in Sponsored Search Advertisements, D. Sculley, Robert Malkin, Sugato Basu, Roberto J. Bayardo, Proc. of the 15th International ACM-SIGKDD Conference on Knowledge Discovery and Data Mining, 2009, pp. 1325-1334.