Please use this identifier to cite or link to this item:
http://hdl.handle.net/10397/94578
DC Field | Value | Language |
---|---|---|
dc.contributor | Department of Industrial and Systems Engineering | en_US |
dc.creator | Yang, A | en_US |
dc.creator | Su, Y | en_US |
dc.creator | Wang, Z | en_US |
dc.creator | Jin, S | en_US |
dc.creator | Ren, J | en_US |
dc.creator | Zhang, X | en_US |
dc.creator | Shen, W | en_US |
dc.creator | Clark, JH | en_US |
dc.date.accessioned | 2022-08-25T01:54:03Z | - |
dc.date.available | 2022-08-25T01:54:03Z | - |
dc.identifier.issn | 1463-9262 | en_US |
dc.identifier.uri | http://hdl.handle.net/10397/94578 | - |
dc.language.iso | en | en_US |
dc.publisher | Royal Society of Chemistry | en_US |
dc.rights | This journal is © The Royal Society of Chemistry 2021 | en_US |
dc.rights | The following publication Yang, A., Su, Y., Wang, Z., Jin, S., Ren, J., Zhang, X., ... & Clark, J. H. (2021). A multi-task deep learning neural network for predicting flammability-related properties from molecular structures. Green Chemistry, 23(12), 4451-4465 is available at https://doi.org/10.1039/D1GC00331C. | en_US |
dc.title | A multi-task deep learning neural network for predicting flammability-related properties from molecular structures | en_US |
dc.type | Journal/Magazine Article | en_US |
dc.description.otherinformation | Title on author’s file: Multi-task deep neural network for predicting flammability-related properties: learning from molecular graphs | en_US |
dc.identifier.spage | 4451 | en_US |
dc.identifier.epage | 4465 | en_US |
dc.identifier.volume | 23 | en_US |
dc.identifier.issue | 12 | en_US |
dc.identifier.doi | 10.1039/d1gc00331c | en_US |
dcterms.abstract | It is significant that hazardous properties of chemicals including replacements for banned or restricted products are assessed at an early stage of product and process design. This work proposes a new strategy of modeling quantitate structure-property relationships based on multi-task deep learning for simultaneously predicting four flammability-related properties including lower and upper flammable limits, auto-ignition point temperature and flash point temperature. A multi-task deep neural network (MDNN) has been developed to extract molecular features automatically and correlate multiple properties integrating a Tree-LSTM neural network with multiple feedforward neural networks. Molecular features are encoded in molecular tree graphs, calculated and extracted without manual actions of the user or preliminary molecular descriptor calculation. Two methods, joint training and alternative training, were both employed to train the proposed MDNN, which could capture the relevant information and commonality among multiple target properties. The outlier detection and determination of applicability domain were also introduced into the evaluation of deep learning models. Since the proposed MDNN utilized data more efficiently, the finally obtained model performs better than the multi-task partial least squares model on predicting the flammability-related properties. The proposed framework of multi-task deep learning provides a promising tool to predict multiple properties without calculating descriptors. | en_US |
dcterms.accessRights | open access | en_US |
dcterms.bibliographicCitation | Green chemistry, 21 June 2021, v. 23, no. 12, p. 4451-4465 | en_US |
dcterms.isPartOf | Green chemistry | en_US |
dcterms.issued | 2021-06-21 | - |
dc.identifier.scopus | 2-s2.0-85108536427 | - |
dc.description.validate | 202208 bcww | en_US |
dc.description.oa | Accepted Manuscript | en_US |
dc.identifier.FolderNumber | ISE-0119 | - |
dc.description.fundingSource | Others | en_US |
dc.description.fundingText | National Natural Science Foundation of China; Beijing Hundreds of Leading Talents Training Project of Science and Technology; Joint Supervision Scheme with the Chinese Mainland, Taiwan and Macao Universities – Other Chinese Mainland, Taiwan and Macao Universities | en_US |
dc.description.pubStatus | Published | en_US |
dc.identifier.OPUS | 53034920 | - |
dc.description.oaCategory | Green (AAM) | en_US |
Appears in Collections: | Journal/Magazine Article |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Yang_Multi-task_Deep_Learning.pdf | Pre-Published version | 2.49 MB | Adobe PDF | View/Open |
Page views
98
Last Week
0
0
Last month
Citations as of Sep 22, 2024
Downloads
132
Citations as of Sep 22, 2024
SCOPUSTM
Citations
15
Citations as of Sep 26, 2024
WEB OF SCIENCETM
Citations
14
Citations as of Sep 26, 2024
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.