Software Tools for ROC and AUC Estimates

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62690094%3A18450%2F03%3A00000620" target="_blank" >RIV/62690094:18450/03:00000620 - isvavai.cz</a>

Result on the web

—

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Software Tools for ROC and AUC Estimates

Original language description

Discrimination and classification are important tasks that enable to solve the problem of allocation objects into one of the predefined classes. This paper concentrates on the problem of discrimination between two populations (groups) of objects. ROC curve is recommended here as a meas-ure of the separation ability of different classification models when conditions of classification change. ROC can be used for description of predictive properties of different classification models. An overview and comparison of the software tools for ROC and AUC analysis are presented here. Results from analysis of the performance of different models for good and bad loans prediction (data for simulated - fictive bank) are used as an illustrative example.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

BB - Applied statistics, operational research

OECD FORD branch

—

Project

—

Continuities

Z - Vyzkumny zamer (s odkazem do CEZ)

Publication year

2003

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 21th International Conference Mathematical Methods in Economics 2003

ISBN

80-213-1046-4

ISSN

—

e-ISSN

—

Number of pages

6

Pages from-to

238-243

Publisher name

Czech University of Agriculture in Prague

Place of publication

Praha

Event location

Praha

Event date

—

Type of event by nationality

—

UT code for WoS article

—