Prediction of absolute entropy of ideal gas at 298K of pure chemicals through GAMLR and FFNN

Multi‐expression programming based model for prediction of formation enthalpies of nitro‐energetic materials

“…In order to obtain a consistent data division, several combinations of the training and testing sets were considered. The selection was such that the maximum, minimum, mean and standard deviations of parameters were consistent in training and testing data sets (Fazeli et al ., ). For the analysis, 28 values (80%) of the data are taken for training and the rest of the values (20%) for testing the generalization capability of the models.…”

“…In splitting the original data set into training and test sets, the composition of the two sets is of crucial importance. The best splitting must guarantee that the training and sets are scattered over the whole area occupied by representative points in the descriptor space (reappearance) and that the training set is distributed over the area occupied by representative points for the whole data set (diversity) (Gharagheizi et al ., ; Godavarthy et al ., ; Katritzky et al ., ; Mehrpooya & Gharagheizi, ; Pan et al ., ; Bagheri et al ., , , ; Fazeli et al ., ; Gandomi et al ., ).…”

“…The development of a QSPR model involves the following series of steps (Gharagheizi et al ., ; Godavarthy et al ., ; Katritzky et al ., ; Pan et al ., ; Mehrpooya & Gharagheizi, ; Bagheri et al ., , , ; Fazeli et al ., ): (a) data set preparation, (b) descriptor calculation, (c) model developing, (d) model validation, and (e) model interpretation.…”

“…Another problem is the lack of measured data available to extend these methods to less‐common energetic materials such as molecules containing fused aromatic rings or polyfunctional substitutions. Last but not the least, the availability of different group contribution based methods is of questionable accuracy and precludes their applicability especially for the nitro‐energetic materials (Gharagheizi et al ., ; Godavarthy et al ., ; Katritzky et al ., ; Mehrpooya & Gharagheizi, ; Pan et al ., ; Bagheri et al ., ; ; ; Fazeli et al ., ). The above‐mentioned limitations demonstrate a need for a different modelling framework, such as that encompassed by QSPR. This technique is based on molecular structure, which not only allows for a fundamental modelling approach, but also allows for a more transparent interpretation of the structural relationship to a physical phenomenon.…”

“…One of the most important aspects in QSPR modelling is the selection of adequate descriptors containing the information stored in the molecular structure. Thus, one can conclude that, as the theoretical descriptors are solely derived from the molecular structure, the QSPR should, in principle, be applicable to any chemical structure (Gharagheizi et al ., ; Godavarthy et al ., ; Katritzky et al ., ; Mehrpooya & Gharagheizi, ; Pan et al ., ; Bagheri et al ., , , ; Fazeli et al ., ).…”