Measurement and ANN prediction of pH-dependent solubility of nitrogen-heterocyclic compounds

Sun, Feifei; Yu, Qihui; Zhu, Jun; Lei, Lecheng; Li, Zhongjian; Zhang, Qizhou

doi:10.1016/j.chemosphere.2015.04.092

Cited by 14 publications

(6 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this kind of FFNN, the input layer receives data from the outside and feeds the network. The output layer represents the outputs and measured solubility 38 . The input and output layers have the same number of neurons as the input and output attributes.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Investigation and feed‐forward neural network‐based estimation of dyeing properties of air plasma treated wool fabric dyed with natural dye obtained from Hibiscus sabdariffa

Başyiğit

Eyüpoğlu

et al. 2023

Coloration Technology

View full text Add to dashboard Cite

In the colouring processes of textile products, more environmentally friendly chemicals and finishing methods should be used instead of conventional ones that harm the environment every day, so that alternative realistic ways to protect nature, both academically and industrially, could be possible. Due to some inconveniences caused by synthetic dyes that are widely used today, in this study, ultrasonic dyeing of wool fabric with Hibiscus sabdariffa was carried out after environmental‐friendly air vacuum plasma application which increased the absorption of the dyes into the textile material. According to the performance results, colour strengths of the wool fabrics were increased significantly. Surface morphology analysis was carried out and etching effects of air vacuum plasma treatment were clearly seen on the micrographs of the treated wool fabrics. An environmental‐friendly green process was achieved through this study and it was concluded that vacuum air plasma treatment could be an alternative green‐process as a pretreatment to increase the dye up‐take of natural dyeing treatment. Moreover, in this study, a feed‐forward neural network (FFNN) model was presented and used for predicting the dyeing properties (L, a, b and K/S) of samples. The experimental results showed that the presented model achieves the regression values greater than 0.9 for all dyeing properties. Consequently, it was considered that the proposed FFNN was successfully modelled and could be efficiently utilised for dyeing characteristics of wool fabrics dyed with natural dye extracted from Hibiscus sabdariffa.

show abstract

Section: Methodsmentioning

confidence: 99%

“…The output layer represents the outputs and measured solubility. 38 The input and output layers have the same number of neurons as the input and output attributes. The hidden layer is used to optimise the network.…”

Section: Feed-forward Neural Network Modellingmentioning

confidence: 99%

Investigation and feed‐forward neural network‐based estimation of dyeing properties of air plasma treated wool fabric dyed with natural dye obtained from Hibiscus sabdariffa

Başyiğit

Eyüpoğlu

et al. 2023

Coloration Technology

View full text Add to dashboard Cite

show abstract

“…In FFNN models, the input layer is used for receiving data from outside and feeding the network, while the output layer presents the outputs and measured solubility. 26,30 The number of neurons in each of the input and output layers is the same as each of the input and output attributes counted. The network is optimised by using the hidden layer.…”

Section: Proposed Ann Modelmentioning

confidence: 99%

Natural dyeing of air plasma‐treated wool fabric with Rubia tinctorum L. and prediction of dyeing properties using an artificial neural network

Eyüpoğlu

Merdan

et al. 2023

Coloration Technology

View full text Add to dashboard Cite

In this study, the ecological dyeing process of wool fabrics was investigated. Wool fabric samples were treated with atmospheric pressure plasma‐jet and corona discharge plasma to modify the surface to make the process sustainable and greener. The samples were dyed with the aqueous extract procured from the powdered roots of Rubia tinctorum L. (madder) using the ultrasonic‐assisted method. Scanning electron microscopy and Fourier Transform–infrared analysis were performed to investigate the effect of plasma treatment on the physical and chemical properties of wool fibres. The effects of plasma treatment type, plasma treatment parameters and the duration of dyeing on colorimetric and fastness properties were investigated. The etching of the wool fibre surface and roughness after plasma treatment were proven with scanning electron microscopy images. The Fourier Transform–infrared spectra showed that no significant differences in the functional groups of wool fibre occurred after plasma treatment. The experimental results proved that plasma treatment parameters and dyeing time had an effect on the colorimetric and fastness properties of the samples. Furthermore, an artificial neural network model was proposed for estimating the dyeing properties of wool fabrics, namely, L, a, b, K/S, colour change, rubbing fastness (dry) and rubbing fastness (wet). The experimental results show that the proposed model achieves regression values greater than 0.97 for all dyeing properties. The proposed model is successful and can be efficiently used for estimating the dyeing characteristics of wool fabrics.

show abstract

“…A validation set [21] was created by examining the SVM output of the 359 small-molecule drugs from the known BCS classification as recognized by the World Health…”

Section: Svm Verification Of Bcs Classificationmentioning

confidence: 99%

Simulation-based comparison of Biopharmaceutics Classification System and drug structure

Chen

et al. 2019

Preprint

View full text Add to dashboard Cite

Background:The Biopharmaceutics Classification System (BCS), which classifies bioactive molecules based on solubility and permeability, is widely used to guide new drug development and drug formulation, as well as predict pharmacokinetics. Here we performed computer simulations to study correlations between a molecule's structure and its BCS classification.Methods:A total of 411 small molecules were assigned to BCS categories based on published drug data, and their Pybel-FP4 fingerprints were extrapolated. The information gain(IG) of each fingerprint was calculated and its characteristic structure analyzed. IG was calculated using multiple thresholds, and results were verified using support vector machine prediction, while taking into account the dose coefficient(0-0.1, 0.1-1, or>1). Structural functional features common to fingerprints of compounds in each type of BCS class were determined using computer simulations.Results:BCS classes III and IV appear to share several structural and functional characteristics, including Secondary aliphaticamine, Michael_acceptor, Isothiourea, and Sulfonamide Sulfonic_derivatives. Conclusion:We demonstrate that our approach can correlate characteristic fingerprints of small-molecule drugs with BCS classifications, which may help guide the development and optimization of new drugs.

show abstract

Measurement and ANN prediction of pH-dependent solubility of nitrogen-heterocyclic compounds

Cited by 14 publications

References 28 publications

Investigation and feed‐forward neural network‐based estimation of dyeing properties of air plasma treated wool fabric dyed with natural dye obtained from Hibiscus sabdariffa

Investigation and feed‐forward neural network‐based estimation of dyeing properties of air plasma treated wool fabric dyed with natural dye obtained from Hibiscus sabdariffa

Natural dyeing of air plasma‐treated wool fabric with Rubia tinctorum L. and prediction of dyeing properties using an artificial neural network

Simulation-based comparison of Biopharmaceutics Classification System and drug structure

Contact Info

Product

Resources

About