Increasing the performance, trustworthiness and practical value of machine learning models: a case study predicting hydrogen bond network dimensionalities from molecular diagrams-Reference-Cited by-同舟云学术

Increasing the performance, trustworthiness and practical value of machine learning models: a case study predicting hydrogen bond network dimensionalities from molecular diagrams

Published:2020 Issue:43 Volume:22 Page:7186-7192
ISSN:1466-8033
Container-title:CrystEngComm
language:en
Short-container-title:CrystEngComm

Author:

Frade Andre P.¹²³⁴^ORCID,McCabe Patrick⁵⁶⁴^ORCID,Cooper Richard I.¹²³⁴^ORCID

Affiliation:

1. Chemical Crystallography Laboratory

2. Department of Chemistry

3. University of Oxford

4. UK

5. Cambridge Crystallographic Data Centre

6. Cambridge

Abstract

The value of a hydrogen bond network prediction model was improved using a tool to increase prediction trust. Its accuracy could be improved up to 73% or 89% with the compromise that only 34% and 8% of the test examples could be predicted.

Funder

Engineering and Physical Sciences Research Council

University of Oxford

Pfizer UK

Medical Research Council

Publisher

Royal Society of Chemistry (RSC)

Subject

Condensed Matter Physics,General Materials Science,General Chemistry

Link

http://pubs.rsc.org/en/content/articlepdf/2020/CE/D0CE00111B

Reference26 articles.