NEURAL-NETWORK - TOPOLOGICAL INDEXES APPROACH TO THE PREDICTION OF PROPERTIES OF ALKENE

Authors
LIN SH ZHANG RS LIU MC HU ZD
Citation
Sh. Lin et al., NEURAL-NETWORK - TOPOLOGICAL INDEXES APPROACH TO THE PREDICTION OF PROPERTIES OF ALKENE, Journal of chemical information and computer sciences, 37(6), 1997, pp. 1146-1151
Citations number
12
Categorie Soggetti
Information Science & Library Science","Computer Application, Chemistry & Engineering","Computer Science Interdisciplinary Applications",Chemistry,"Computer Science Information Systems
Journal title
Journal of chemical information and computer sciences → ACNP
ISSN journal
00952338
Volume
37
Issue
6
Year of publication
1997
Pages
1146 - 1151
Database
ISI
SICI code
0095-2338(1997)37:6<1146:N-TIAT>2.0.ZU;2-G
Abstract
A topological indices vector of five parameters (chi, P, w, l, s) incl uding three grades of structural information was set up as a molecular descriptor to predict the normal boiling point, the density, and the refractive index of alkenes with a neural network. The five parameters are the connection index chi, the polarity number p, w, l representin g the effect of a double bond on the properties, and s distinguishing enantiomers of alkenes. The estimation results show average accuracies of 1.3% with maximum deviations of 16%.