Research Interests:
Managing storage in the face of relentless growth in the number and va- riety of files on storage systems creates demand for rich file s ystem meta- data as is made evident by the recent emergence of rich metadata support in many... more
Managing storage in the face of relentless growth in the number and va- riety of files on storage systems creates demand for rich file s ystem meta- data as is made evident by the recent emergence of rich metadata support in many applications as well as file systems. Yet, little suppor t exists for shar- ing metadata across file systems
Research Interests:
Although boosting methods have become an extremely important classification method, there has been little attention paid to boosting with asymmetric losses. In this paper we take a gradient descent view of boosting in order to motivate a... more
Although boosting methods have become an extremely important classification method, there has been little attention paid to boosting with asymmetric losses. In this paper we take a gradient descent view of boosting in order to motivate a new boosting variant called BiBoost which treats the two classes differently. This variant is likely to perform well when there is a different cost for false positive and false negative predic-tions. The variant is also appropriate when the data comes from multiple sources with different reliabilities or noise levels. Experiments show that BiBoost effectively reduces the number of false positive mistakes, and a more general algorithm is discussed.