Document Detail

Analysis of linear and nonlinear QSAR data using neural networks.
MedLine Citation:
PMID:  7966135     Owner:  NLM     Status:  MEDLINE    
The use of feed forward back propagation neural networks to perform the equivalent of multiple linear regression has been examined using artificial structured data sets and real literature data. Their predictive ability has been assessed using leave-one-out cross-validation and training/test set protocols. While networks have been shown to fit data sets well, they appear to suffer from a number of disadvantages. In particular, they have performed poorly in prediction for the QSAR data examined here, they are susceptible to chance effects, and the relationships developed by the networks are difficult to interpret. This investigation reports results for one particular form of artificial neural network; other architectures and applications, however, may be more suitable.
D T Manallack; D D Ellis; D J Livingstone
Related Documents :
18244505 - Gensofnn: a generic self-organizing fuzzy neural network.
17873435 - Free-paced high-performance brain-computer interfaces.
11543265 - Multivariable empirical modeling of als systems using polynomials.
16687235 - A neural network implementation of a saliency map model.
20733235 - Model synthesis: a general procedural modeling algorithm.
22373375 - User centered and ontology based information retrieval system for life sciences.
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  Journal of medicinal chemistry     Volume:  37     ISSN:  0022-2623     ISO Abbreviation:  J. Med. Chem.     Publication Date:  1994 Oct 
Date Detail:
Created Date:  1994-12-20     Completed Date:  1994-12-20     Revised Date:  2000-12-18    
Medline Journal Info:
Nlm Unique ID:  9716531     Medline TA:  J Med Chem     Country:  UNITED STATES    
Other Details:
Languages:  eng     Pagination:  3758-67     Citation Subset:  IM    
SmithKline Beecham Pharmaceuticals, Welwyn, Herts, U.K.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Neural Networks (Computer)*
Structure-Activity Relationship*

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Previous Document:  Generally applicable, convenient solid-phase synthesis and receptor affinities of octreotide analogs...
Next Document:  A generalized formalism of three-dimensional quantitative structure-property relationship analysis f...