JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 378
Type: Contributed
Date/Time: Tuesday, August 11, 2015 : 11:35 AM to 12:20 PM
Sponsor: Section on Statistical Learning and Data Mining
Abstract #317888
Title: Binormal ROC and Precision-Recall Classification with Nonparametric Functions
Author(s): Yingzi Xu* and Howard Bondell
Companies: North Carolina State University and North Carolina State University
Keywords: Binary classification ; Precision-Recall curve ; ROC curve ; Radial basis function ; B-spline ; Binormal assumption
Abstract:

Both Precision-Recall (PR) curve and Receiver Operating Characteristic (ROC) curve are highly informative about the performance of binary classifiers, and the area under these curves are very popular for comparing different classifiers. In this paper, we propose a novel approach for binary classification based on maximizing the area under the PR curve and the ROC curve under the binormal assumption. The key idea is to estimate the optimal classifier by maximizing the area under PR curve with a gradient descent algorithm or by maximizing the area under ROC curve with a closed form derived. It utilizes non-parametric functions, e.g. radial basis functions and b-splines to approximate the true function which overcomes the fully parametric assumption used in linear classifiers.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home