TY - GEN

T1 - Node Classification with Bounded Error Rates

AU - Wijegunawardana, Pivithuru

AU - Gera, Ralucca

AU - Soundarajan, Sucheta

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Node classification algorithms are widely used for the task of node label prediction in partially labeled graph data. In many problems, a user may wish to associate a confidence level with a prediction such that the error in the prediction is guaranteed. We propose adopting the Conformal Prediction framework [17] to obtain guaranteed error bounds in node classification problem. We show how this framework can be applied to (1) obtain predictions with guaranteed error bounds, and (2) improve the accuracy of the prediction algorithms. Our experimental results show that the Conformal Prediction framework can provide up to a 30% improvement in node classification algorithm accuracy while maintaining guaranteed error bounds on predictions.

AB - Node classification algorithms are widely used for the task of node label prediction in partially labeled graph data. In many problems, a user may wish to associate a confidence level with a prediction such that the error in the prediction is guaranteed. We propose adopting the Conformal Prediction framework [17] to obtain guaranteed error bounds in node classification problem. We show how this framework can be applied to (1) obtain predictions with guaranteed error bounds, and (2) improve the accuracy of the prediction algorithms. Our experimental results show that the Conformal Prediction framework can provide up to a 30% improvement in node classification algorithm accuracy while maintaining guaranteed error bounds on predictions.

KW - Bounded error rates

KW - Conformal prediction

KW - Node classification

UR - http://www.scopus.com/inward/record.url?scp=85081267097&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85081267097&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-40943-2_3

DO - 10.1007/978-3-030-40943-2_3

M3 - Conference contribution

AN - SCOPUS:85081267097

SN - 9783030409425

T3 - Springer Proceedings in Complexity

SP - 26

EP - 38

BT - Complex Networks XI - Proceedings of the 11th Conference on Complex Networks, CompleNet 2020

A2 - Barbosa, Hugo

A2 - Menezes, Ronaldo

A2 - Gomez-Gardenes, Jesus

A2 - Gonçalves, Bruno

A2 - Mangioni, Giuseppe

A2 - Oliveira, Marcos

PB - Springer

T2 - 11th International Conference on Complex Networks, CompleNet 2020

Y2 - 31 March 2020 through 3 April 2020

ER -