TY - GEN
T1 - Learning using privileged information (LUPI) for modeling survival data
AU - Shiao, Han Tai
AU - Cherkassky, Vladimir
PY - 2014/9/3
Y1 - 2014/9/3
N2 - Survival data is common in medical applications. The challenge in applying predictive data-analytic methods to survival data is in the treatment of censored observations, since the survival times for these observations are unknown. This paper presents formalization of the analysis of survival data as a binary classification problem. For this binary classification setting, we propose a strategy for encoding censored data, leading to the SVM/LUPI formulations. Further, we present empirical comparison of the new method and the classical Cox modeling approach for predictive modeling of survival data. These comparisons suggest that for data sets with large amount of censored data, the proposed method consistently yields better predictive performance than classical statistical modeling.
AB - Survival data is common in medical applications. The challenge in applying predictive data-analytic methods to survival data is in the treatment of censored observations, since the survival times for these observations are unknown. This paper presents formalization of the analysis of survival data as a binary classification problem. For this binary classification setting, we propose a strategy for encoding censored data, leading to the SVM/LUPI formulations. Further, we present empirical comparison of the new method and the classical Cox modeling approach for predictive modeling of survival data. These comparisons suggest that for data sets with large amount of censored data, the proposed method consistently yields better predictive performance than classical statistical modeling.
UR - http://www.scopus.com/inward/record.url?scp=84908474226&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84908474226&partnerID=8YFLogxK
U2 - 10.1109/IJCNN.2014.6889517
DO - 10.1109/IJCNN.2014.6889517
M3 - Conference contribution
AN - SCOPUS:84908474226
T3 - Proceedings of the International Joint Conference on Neural Networks
SP - 1042
EP - 1049
BT - Proceedings of the International Joint Conference on Neural Networks
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2014 International Joint Conference on Neural Networks, IJCNN 2014
Y2 - 6 July 2014 through 11 July 2014
ER -