Multilevel adaptive description of molecular graphs in the "structure-property" problem

Authors:
A. V. Bekker;A. A. Suleimanov;G. N. Apryshko;M. I. Kumskov;R. B. Pugacheva
Affiliations:
Department of Mechanics and Mathematics, Moscow State University, Moscow, Russia 118992;Department of Mechanics and Mathematics, Moscow State University, Moscow, Russia 118992;Department of Mechanics and Mathematics, Moscow State University, Moscow, Russia 118992;Department of Mechanics and Mathematics, Moscow State University, Moscow, Russia 118992;Department of Mechanics and Mathematics, Moscow State University, Moscow, Russia 118992
Venue:
Pattern Recognition and Image Analysis
Year:
2013

Citing 1
Cited 0

A study of cross-validation and bootstrap for accuracy estimation and model selection

IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2

Quantified Score

Hi-index	0.00

Visualization

Abstract

A method for solving the "structure-property" problem is proposed based on the adaptive choice of molecule description and automatic selection of a feature space according to the characteristics of the learning sample. A problem of combinatory explosion is solved using the method for group account of arguments. The algorithms for cluster analysis are used to improve the predictive capability of the model. The results of model construction using specific learning samples of chemical compounds are discussed.