Thors: An Efficient Approach for Making Classifiers Cost-Sensitive

By Ye Tian et al

Read the original document by opening this link in a new tab.

Table of Contents

1 Introduction
2 The THORS Algorithm
3 Properties

Summary

The paper introduces THORS, a method for making classifiers cost-sensitive by using order statistic. It proposes an effective thresholding method based on order statistic to convert scoring-type classifiers into cost-sensitive ones without requiring deep knowledge of the classifiers. The paper discusses theoretical properties, time complexity, and bounds for misclassification costs of THORS. It presents the THORS algorithm and its properties in detail.