A comparison of "neural networks and multiple linear regressions" models to describe the rejection of micropollutants by membranes
A rejection process of organic compounds by nanofiltration and reverse osmosis membranes was modelled using the artificial neural networks. Three feed-forward neural networks based on quantitative structure-activity relationship (QSAR-NN models) characterised by a similar structure (twelve neurons f...
| Permalink: | http://skupni.nsk.hr/Record/nsk.NSK01001098251/Details |
|---|---|
| Matična publikacija: |
Kemija u industriji (Online) 69 (2020), 3/4 ; str. 111-127 |
| Glavni autori: | Ammi, Yamina (Author), Khaouane, Latifa, Hanini, Salah |
| Vrsta građe: | e-članak |
| Jezik: | eng |
| Predmet: | |
| Online pristup: |
https://doi.org/10.15255/KUI.2019.024 Kemija u industriji (Online) Hrčak |
| LEADER | 04044naa a22003854i 4500 | ||
|---|---|---|---|
| 001 | NSK01001098251 | ||
| 003 | HR-ZaNSK | ||
| 005 | 20210614145709.0 | ||
| 006 | m d | ||
| 007 | cr|||||||||||| | ||
| 008 | 210420s2020 ci d |o |0|| ||eng | ||
| 024 | 7 | |2 doi |a 10.15255/KUI.2019.024 | |
| 035 | |a (HR-ZaNSK)001098251 | ||
| 040 | |a HR-ZaNSK |b hrv |c HR-ZaNSK |e ppiak | ||
| 041 | 0 | |a eng |b hrv |b eng | |
| 042 | |a croatica | ||
| 044 | |a ci |c hr | ||
| 080 | 1 | |a 66 |2 2011 | |
| 100 | 1 | |a Ammi, Yamina |4 aut |9 HR-ZaNSK | |
| 245 | 1 | 2 | |a A comparison of "neural networks and multiple linear regressions" models to describe the rejection of micropollutants by membranes |h [Elektronička građa] / |c Yamina Ammi, Latifa Khaouane, Salah Hanini. |
| 300 | |b Graf. prikazi. | ||
| 504 | |a Bibliografija: 53 jed. | ||
| 504 | |a Summary ; Sažetak. | ||
| 520 | |a A rejection process of organic compounds by nanofiltration and reverse osmosis membranes was modelled using the artificial neural networks. Three feed-forward neural networks based on quantitative structure-activity relationship (QSAR-NN models) characterised by a similar structure (twelve neurons for QSAR-NN1, QSAR-NN2, and QSAR-NN3 in the input layer, one hidden layer and one neuron in the output layer), were constructed with the aim of predicting the rejection of organic compounds. A set of 1394 data points for QSAR-NN1, 980 data points for QSAR-NN2, and 436 data points for QSAR-NN3 were used to construct the neural networks. Good agreements between the predicted and experimental rejections were obtained by QSAR-NN models (the correlation coefficient for the total dataset were 0.9191 for QSAR-NN1, 0.9338 for QSAR-NN2, and 0.9709 for QSAR-NN3). Comparison between the feed-forward neural networks and multiple linear regressions based on quantitative structure-activity relationship “QSAR-MLR” revealed the superiority of the QSAR-NN models (the root mean squared errors for the total dataset for the QSAR-NN models were 10.6517 % for QSAR-NN1, 9.1991 % for QSAR-NN2, and 5.8869 % for QSAR-NN3, and for QSAR-MLR models they were 20.1865 % for QSAR-MLR1, 19.3815 % for QSAR-MLR2, and 16.2062 % for QSAR-MLR3). | ||
| 520 | |a Postupak odbacivanja organskih spojeva nanofiltracijom i membranama reverzne osmoze modeliran je umjetnim neuronskim mrežama. Konstruirane su tri neuronske mreže zasnovane na kvantitativnom odnosu strukture-aktivnosti (QSAR-NN modeli) karakterizirane sličnom strukturom (dvanaest neurona za QSAR-NN1, QSAR-NN2 i QSAR-NN3 u ulaznom sloju, jedan skriveni sloj i jedan neuron u izlaznom sloju), s ciljem predviđanja odbacivanja organskih spojeva. Za izgradnju neuronskih mreža upotrijebljeni su skupovi od 1394 podatkovnih točaka za QSAR-NN1, 980 podatkovnih točaka za QSAR-NN2 i 436 podatkovnih točaka za QSAR-NN3. Dobre usklađenosti između predviđenih i eksperimentalnih odbacivanja dobivene su modelima QSAR-NN (korelacijski koeficijent za ukupni skup podataka bio je 0,9191 za QSAR-NN1, 0,9338 za QSAR-NN2 i 0,9709 za QSAR-NN3). Usporedba neuronskih mreža i višestrukih linearnih regresija zasnovanih na kvantitativnom odnosu struktura-aktivnost “QSAR-MLR” otkrila je superiornost modela QSAR-NN (korijenske srednje kvadratne pogreške za ukupni skup podataka za modele QSAR-NN bile su 10,6517 % za QSAR-NN1, 9,1991 % za QSAR-NN2, i 5,8869 % za QSAR-NN3, a za modele QSAR-MLR 20,1865 % za QSAR-MLR1, 19,3815 % za QSAR-MLR2, i 16,2062 % za QSAR-MLR3). | ||
| 653 | 0 | |a Neuronske mreže |a Linearna regresija |a Mikroonečišćenje |a Nanofiltracija |a Organski spojevi | |
| 700 | 1 | |a Khaouane, Latifa |4 aut |9 HR-ZaNSK | |
| 700 | 1 | |a Hanini, Salah |4 aut |9 HR-ZaNSK | |
| 773 | 0 | |t Kemija u industriji (Online) |x 1334-9090 |g 69 (2020), 3/4 ; str. 111-127 |w nsk.(HR-ZaNSK)000530475 | |
| 981 | |b Be2020 |b B03/20 | ||
| 998 | |b dalo2106 | ||
| 856 | 4 | 0 | |u https://doi.org/10.15255/KUI.2019.024 |
| 856 | 4 | 0 | |u http://silverstripe.fkit.hr/kui/issue-archive/article/693 |y Kemija u industriji (Online) |
| 856 | 4 | 0 | |u https://hrcak.srce.hr/235865 |y Hrčak |
| 856 | 4 | 1 | |y Digitalna.nsk.hr |