REPOSITORY QDB RESOURCES NEWS CONTACTS

Tetteh, J.; Suzuki, T.; Metcalfe, E.; Howells, S. Quantitative Structure-Property Relationships for the Estimation of Boiling Point and Flash Point Using a Radial Basis Function Neural Network. J. Chem. Inf. Model. 1999, 39, 491-507.

QsarDB Repository

Tetteh, J.; Suzuki, T.; Metcalfe, E.; Howells, S. Quantitative Structure-Property Relationships for the Estimation of Boiling Point and Flash Point Using a Radial Basis Function Neural Network. J. Chem. Inf. Model. 1999, 39, 491-507.

QDB archive DOI: 10.15152/QDB.158   DOWNLOAD

QsarDB content

Property Tb: Boiling point [°C]

Compounds: 400 | Models: 1 | Predictions: 3

Tab7_Tb: RBF-NN model for diverse chemicals i

Neural network (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set training 135 0.973 12.754
Validation set external validation 133 0.957 15.683
Testing set external validation 132 0.915 21.674

Property Tf: Flash point [°C]

Compounds: 400 | Models: 1 | Predictions: 3

Tab7_Tf: RBF-NN model for diverse chemicals i

Neural network (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set training 135 0.963 9.879
Validation set external validation 133 0.928 14.100
Testing set external validation 132 0.922 13.964

Citing

When citing this QDB archive, please also include the original article:

  • Tetteh, J.; Suzuki, T.; Metcalfe, E.; Howells, S. Quantitative Structure-Property Relationships for the Estimation of Boiling Point and Flash Point Using a Radial Basis Function Neural Network. J. Chem. Inf. Model. 1999, 39, 491-507. http://dx.doi.org/10.1021/ci980026y

  • Ahte, P. QDB archive #158. QsarDB repository, 2015. http://dx.doi.org/10.15152/QDB.158

Metadata

Show full item record

Title: Tetteh, J.; Suzuki, T.; Metcalfe, E.; Howells, S. Quantitative Structure-Property Relationships for the Estimation of Boiling Point and Flash Point Using a Radial Basis Function Neural Network. J. Chem. Inf. Model. 1999, 39, 491-507.
Abstract: Radial basis function (RBF) neural network models for the simultaneous estimation of flash point (Tf) and boiling point (Tb) based on 25 molecular functional groups and their first-order molecular connectivity index (1χ) have been developed. The success of the whole modeling process depended on a network optimization strategy based on biharmonic spline interpolation for the selection of an optimum RBF neurons (n) in the hidden layer and their associated spread parameter (σ). The RBF networks were trained by the Orthogonal Least Squares (OLS) learning algorithm. After dividing the total database of 400 compounds into training (134), validation (133), and testing (133), the average absolute errors obtained for the validation and testing sets ranges from 10 °C to 12 °C and 11 °C to14 °C for Tf and Tb, respectively, and are in agreement with the experimental value of about 10 °C. Results of a standard Partial Least Square (PLS) regression model for single output predictions range from 23 °C to 24 °C and 18 °C to 20 °C for Tf and Tb, respectively, indicating the superior predictive ability of the neural model and strongly suggests that a nonlinear relationship exists between the input and target parameters of the data. The robustness of the neural model was successfully examined by a random split cross validation based on pooling together of the validation and test data sets. The study shows that the simple numerical coding of a molecule based on its formula together with its 1χ is an attractive way of estimating the flammability properties of organic compounds via an RBF neural network
URI: http://hdl.handle.net/10967/158
http://dx.doi.org/10.15152/QDB.158
Date: 2015-05-28


Files in this item

Name Description Format Size View
1999JCIM491.qdb.zip RBF-NN models for boiling point and flash point application/x-zip 126.4Kb View/Open
Files associated with this item are distributed
under Creative Commons license.

This item appears in the following Collection(s)

Show full item record