A focus on the use of real-world datasets for yield prediction

The prediction of reaction yields remains a challenging task for machine learning (ML), given the vast search spaces and absence of robust training data. Wiest, Chawla et al. (https://doi.org/10.1039/D2SC06041H) show that a deep learning algorithm performs well on high-throughput experimentation dat...

Descripción completa

Detalles Bibliográficos
Autores principales: Bustillo, Latimah, Rodrigues, Tiago
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Royal Society of Chemistry 2023
Materias:
Chemistry
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10189867/
https://www.ncbi.nlm.nih.gov/pubmed/37206402
http://dx.doi.org/10.1039/d3sc90069j
_version_ 1785043174528385024
author Bustillo, Latimah
Rodrigues, Tiago
author_facet Bustillo, Latimah
Rodrigues, Tiago
author_sort Bustillo, Latimah
collection PubMed
description The prediction of reaction yields remains a challenging task for machine learning (ML), given the vast search spaces and absence of robust training data. Wiest, Chawla et al. (https://doi.org/10.1039/D2SC06041H) show that a deep learning algorithm performs well on high-throughput experimentation data but surprisingly poorly on real-world, historical data from a pharmaceutical company. The result suggests that there is considerable room for improvement when coupling ML to electronic laboratory notebook data.
format Online
Article
Text
id pubmed-10189867
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher The Royal Society of Chemistry
record_format MEDLINE/PubMed
spelling pubmed-101898672023-05-18 A focus on the use of real-world datasets for yield prediction Bustillo, Latimah Rodrigues, Tiago Chem Sci Chemistry The prediction of reaction yields remains a challenging task for machine learning (ML), given the vast search spaces and absence of robust training data. Wiest, Chawla et al. (https://doi.org/10.1039/D2SC06041H) show that a deep learning algorithm performs well on high-throughput experimentation data but surprisingly poorly on real-world, historical data from a pharmaceutical company. The result suggests that there is considerable room for improvement when coupling ML to electronic laboratory notebook data. The Royal Society of Chemistry 2023-04-27 /pmc/articles/PMC10189867/ /pubmed/37206402 http://dx.doi.org/10.1039/d3sc90069j Text en This journal is © The Royal Society of Chemistry https://creativecommons.org/licenses/by/3.0/
spellingShingle Chemistry
Bustillo, Latimah
Rodrigues, Tiago
A focus on the use of real-world datasets for yield prediction
title A focus on the use of real-world datasets for yield prediction
title_full A focus on the use of real-world datasets for yield prediction
title_fullStr A focus on the use of real-world datasets for yield prediction
title_full_unstemmed A focus on the use of real-world datasets for yield prediction
title_short A focus on the use of real-world datasets for yield prediction
title_sort focus on the use of real-world datasets for yield prediction
topic Chemistry
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10189867/
https://www.ncbi.nlm.nih.gov/pubmed/37206402
http://dx.doi.org/10.1039/d3sc90069j
work_keys_str_mv AT bustillolatimah afocusontheuseofrealworlddatasetsforyieldprediction
AT rodriguestiago afocusontheuseofrealworlddatasetsforyieldprediction
AT bustillolatimah focusontheuseofrealworlddatasetsforyieldprediction
AT rodriguestiago focusontheuseofrealworlddatasetsforyieldprediction

Ejemplares similares