P
Paul S. Bradley
Researcher at Microsoft
Publications - 46
Citations - 7057
Paul S. Bradley is an academic researcher from Microsoft. The author has contributed to research in topics: Cluster analysis & Canopy clustering algorithm. The author has an hindex of 28, co-authored 46 publications receiving 6805 citations. Previous affiliations of Paul S. Bradley include EMC Corporation & University of Wisconsin-Madison.
Papers
More filters
Proceedings Article
Refining Initial Points for K-Means Clustering
Paul S. Bradley,Usama M. Fayyad +1 more
TL;DR: A procedure for computing a refined starting condition from a given initial one that is based on an efficient technique for estimating the modes of a distribution that allows the iterative algorithm to converge to a “better” local minimum.
Proceedings Article
Feature Selection via Concave Minimization and Support Vector Machines
TL;DR: Numerical tests on 6 public data sets show that classi ers trained by the concave minimization approach and those trained by a support vector machine have comparable 10fold cross-validation correctness.
Proceedings Article
Scaling clustering algorithms to large databases
TL;DR: A scalable clustering framework applicable to a wide class of iterative clustering that requires at most one scan of the database and is instantiated and numerically justified with the popular K-Means clustering algorithm.
Patent
Apparatus and accompanying methods for visualizing clusters of data and hierarchical cluster classifications
TL;DR: In this article, the authors present an interactive graphical user interface for visualizing clusters (categories) and segments (summarized clusters) of data, which automatically categorizes incoming case data into clusters, summarizes those clusters into segments, determines similarity measures for the segments, scores the selected segments through the similarity measures, and then forms and visually depicts hierarchical organizations of those selected clusters.
Journal ArticleDOI
k-Plane Clustering
TL;DR: A finite new algorithm is proposed for clustering m given points in n-dimensional real space into k clusters by generating k planes that constitute a local solution to the nonconvex problem of minimizing the sum of squares of the 2-norm distances between each point and a nearest plane.