Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

12

ML
2000
ACM

favoriteEmaildiscussreport

105views Machine Learning» more ML 2000»

Multiple Comparisons in Induction Algorithms

13 years 4 months ago

Multiple Comparisons in Induction Algorithms

Download kdl.cs.umass.edu

Abstract. A single mechanism is responsible for three pathologies of induction algorithms: attribute selection errors, overfitting, and oversearching. In each pathology, induction algorithms compare multiple items based on scores from an evaluation function and select the item with the maximum score. We call this a multiple comparison procedure (MCP ). We analyze the statistical properties of MCP s and show how failure to adjust for these properties leads to the pathologies. We also discuss approaches that can control pathological behavior, including Bonferroni adjustment, randomization testing, and cross-validation.

David D. Jensen, Paul R. Cohen

Real-time Traffic

Induction Algorithms | Machine Learning | ML 2000 | Multiple Comparison Procedure | Single Mechanism |

claim paper

Related Content

» Learning Rules for LargeVocabulary Word Sense Disambiguation A Comparison of Various Class...

» FIbased file access predictor

» Induction in Noisy Domains

» Lateen EM Unsupervised Training with Multiple Objectives Applied to Dependency Grammar Ind...

» Project scheduling with multiple modes A comparison of exact algorithms

» Bayesian Comparison of Machine Learning Algorithms on Single and Multiple Datasets

» Faster Algorithms for Optimal Multiple Sequence Alignment Based on Pairwise Comparisons

» A Comparison of Three Rounding Algorithms for IEEE FloatingPoint Multiplication

» Inductive Learning in Less Than One Sequential Data Scan

Post Info
More Details (n/a)

Added	19 Dec 2010
Updated	19 Dec 2010
Type	Journal
Year	2000
Where	ML
Authors	David D. Jensen, Paul R. Cohen

Comments (0)