Abstract
In the last few years, neural networks have found increasing use in chemical applications, including their use in the analysis of quantitative structure‐activity relationships (QSAR) data. Networks are able to perform the equivalent of discriminant and regression analyses, in addition to providing a novel method for the low‐dimensional display of multivariate data. Experiments in our laboratories, using artificially structured data sets and real literature QSAR data with neural networks performing multiple linear regression, demonstrated their susceptibility to over‐fitting, resulting in poor predicted abilities. Other network algorithms and training regimes are emerging in the literature which address these particular problems.
| Original language | English |
|---|---|
| Pages (from-to) | 167-170 |
| Number of pages | 4 |
| Journal | Pesticide Science |
| Volume | 45 |
| Issue number | 2 |
| DOIs | |
| Publication status | Published - 1 Jan 1995 |
| Externally published | Yes |
Keywords
- chance effects
- multiple linear regression
- neutral networks
- QSAR
Cite this
}
Relating biological activity to chemical structure using neural networks. / Manallack, David T.; Livingstone, David J.
In: Pesticide Science, Vol. 45, No. 2, 01.01.1995, p. 167-170.Research output: Contribution to journal › Review Article › Research › peer-review
TY - JOUR
T1 - Relating biological activity to chemical structure using neural networks
AU - Manallack, David T.
AU - Livingstone, David J.
PY - 1995/1/1
Y1 - 1995/1/1
N2 - In the last few years, neural networks have found increasing use in chemical applications, including their use in the analysis of quantitative structure‐activity relationships (QSAR) data. Networks are able to perform the equivalent of discriminant and regression analyses, in addition to providing a novel method for the low‐dimensional display of multivariate data. Experiments in our laboratories, using artificially structured data sets and real literature QSAR data with neural networks performing multiple linear regression, demonstrated their susceptibility to over‐fitting, resulting in poor predicted abilities. Other network algorithms and training regimes are emerging in the literature which address these particular problems.
AB - In the last few years, neural networks have found increasing use in chemical applications, including their use in the analysis of quantitative structure‐activity relationships (QSAR) data. Networks are able to perform the equivalent of discriminant and regression analyses, in addition to providing a novel method for the low‐dimensional display of multivariate data. Experiments in our laboratories, using artificially structured data sets and real literature QSAR data with neural networks performing multiple linear regression, demonstrated their susceptibility to over‐fitting, resulting in poor predicted abilities. Other network algorithms and training regimes are emerging in the literature which address these particular problems.
KW - chance effects
KW - multiple linear regression
KW - neutral networks
KW - QSAR
UR - http://www.scopus.com/inward/record.url?scp=84995012805&partnerID=8YFLogxK
U2 - 10.1002/ps.2780450211
DO - 10.1002/ps.2780450211
M3 - Review Article
VL - 45
SP - 167
EP - 170
JO - Pest Management Science
JF - Pest Management Science
SN - 1526-498X
IS - 2
ER -