Characterizing Cell-Decomposable Metrics

Optimal Decision-Theoretic Classification Using Non-Decomposable Performance Metrics

We provide a general theoretical analysis of expected out-of-sample utility, also referred to as decisiontheoretic classification, for non-decomposable binary classification metrics such as F-measure and Jaccard coefficient. Our key result is that the expected out-of-sample utility for many performance metrics is provably optimized by a classifier which is equivalent to a signed thresholding of...

Consistent Classification Algorithms for Multi-class Non-Decomposable Performance Metrics

We study consistency of learning algorithms for a multi-class performance metric that is anon-decomposable function of the confusion matrix of a classifier and cannot be expressed asa sum of losses on individual data points; examples of such performance metrics include themicro and macro F-measure used widely in information retrieval and the multi-class G-meanmetric popular in c...

Characterizations of Decomposable Dependency Models

Decomposable dependency models possess a number of interesting and useful properties. This paper presents new characterizations of decomposable models in terms of independence relationships, which are obtained by adding a single axiom to the well-known set characterizing dependency models that are isomorphic to undirected graphs. We also brieey discuss a potential application of our results to ...

On-line arbitrarily vertex decomposable trees

A tree T is arbitrarily vertex decomposable if for any sequence of positive integers adding up to the order of T there is a sequence of vertex-disjoint subtrees of T whose orders are given by . An on-line version of the problem of characterizing arbitrarily vertex decomposable trees is completely solved here. © 2007 Elsevier B.V. All rights reserved.

Structure and Applications of Totally Decomposable Metrics