Napredna pretraga

Pregled bibliografske jedinice broj: 768609

Development of multi ANN model for QSRR in ion chromatography


Ukić, Šime; Novak Stankov, Mirjana; Rogošić, Marko; Bolanča, Tomislav
Development of multi ANN model for QSRR in ion chromatography // 21st International Symposium on Separation Sciences, Book of Abstracts / Vovk, Irena ; Glavnik, Vesna ; Albreht, Alen (ur.).
Ljubljana: National Institute of Chemistry, 2015. str. 194-194 (poster, nije recenziran, sažetak, znanstveni)


Naslov
Development of multi ANN model for QSRR in ion chromatography

Autori
Ukić, Šime ; Novak Stankov, Mirjana ; Rogošić, Marko ; Bolanča, Tomislav

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
21st International Symposium on Separation Sciences, Book of Abstracts / Vovk, Irena ; Glavnik, Vesna ; Albreht, Alen - Ljubljana : National Institute of Chemistry, 2015, 194-194

ISBN
978-961-6104-28-9

Skup
21st International Symposium on Separation Sciences

Mjesto i datum
Ljubljana, Slovenija, 30.06.-03.07.2015

Vrsta sudjelovanja
Poster

Vrsta recenzije
Nije recenziran

Ključne riječi
ANN; QSRR; ion chromatography

Sažetak
Nowadays, efficient optimization of separation means much more than simple prediction of retention ; at least peak width and asymmetry should be included while peak shape information would certainly be benefit for global multi-criteria optimization process. However QSRR models prove to be not accurate enough to be able to serve as a core of peak shape model. Therefore, it is essential to raise predictive ability of QSRR model. The aim of this work is development of multi ANN model for QSRR in ion chromatography. Developed model links in parallel up to 100 different ANNs completely trained and optimized for purpose of making single prediction only. The approach discusses influence of number of inbuilt ANN on predictive ability in correlation with time needed for computation. Although results showed significant improvement of predictive ability the straightforward application of developed models for QSRR peak shape modeling still remain questionable.

Izvorni jezik
Engleski

Znanstvena područja
Kemija, Kemijsko inženjerstvo



POVEZANOST RADA


Projekt / tema
110005
125-1253092-3004 - Procesi ionske izmjene u sustavu kvalitete industrijskih voda (Tomislav Bolanča, )

Ustanove
Fakultet kemijskog inženjerstva i tehnologije, Zagreb