Timezone: »

Optimizing F-Measures by Cost-Sensitive Classification
Shameem Puthiya Parambath · Nicolas Usunier · Yves Grandvalet

Wed Dec 10 04:00 PM -- 08:59 PM (PST) @ Level 2, room 210D

We present a theoretical analysis of F-measures for binary, multiclass and multilabel classification. These performance measures are non-linear, but in many scenarios they are pseudo-linear functions of the per-class false negative/false positive rate. Based on this observation, we present a general reduction of F-measure maximization to cost-sensitive classification with unknown costs. We then propose an algorithm with provable guarantees to obtain an approximately optimal classifier for the F-measure by solving a series of cost-sensitive classification problems. The strength of our analysis is to be valid on any dataset and any class of classifiers, extending the existing theoretical results on F-measures, which are asymptotic in nature. We present numerical experiments to illustrate the relative importance of cost asymmetry and thresholding when learning linear classifiers on various F-measure optimization tasks.

Author Information

Shameem Puthiya Parambath (Université de Technologie de Compiègne)
Nicolas Usunier (Université de Technologie de Compiègne (UTC))
Yves Grandvalet (Université de Technologie de Compiégne)

More from the Same Authors