Matches in SemOpenAlex for { <https://semopenalex.org/work/W4253313247> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4253313247 abstract "<p>Inputting molecules into chemistry software, such as quantum chemistry packages, currently requires domain expertise, expensive software and/or cumbersome procedures. Leveraging recent breakthroughs in machine learning, we develop ChemPix: an offline, hand-drawn hydrocarbon structure recognition tool designed to remove these barriers. A neural image captioning approach consisting of a convolutional neural network (CNN) encoder and a long short-term memory (LSTM) decoder learned a mapping from photographs of hand-drawn hydrocarbon structures to machine-readable SMILES representations. We generated a large auxiliary training dataset, based on RDKit molecular images, by combining image augmentation, image degradation and background addition. Additionally, a small dataset of ~600 hand-drawn hydrocarbon chemical structures was crowd-sourced using a phone web application. These datasets were used to train the image-to-SMILES neural network with the goal of maximizing the hand-drawn hydrocarbon recognition accuracy. By forming a committee of the trained neural networks, we achieved a nearly 10 percentage point improvement of the molecule recognition accuracy and were able to assign a confidence value for the prediction based on the number of agreeing votes. The top ensemble model achieved a hand-drawn hydrocarbon recognition accuracy of 77% for the first prediction and 86% if the top 3 predictions were considered; in over 50% of cases, the model was at least 97% confident in the prediction, making it a promising tool for real-world use cases.</p>" @default.
- W4253313247 created "2022-05-12" @default.
- W4253313247 creator A5003577701 @default.
- W4253313247 creator A5040462364 @default.
- W4253313247 creator A5057261669 @default.
- W4253313247 creator A5057461346 @default.
- W4253313247 creator A5062125510 @default.
- W4253313247 creator A5064472074 @default.
- W4253313247 date "2021-03-08" @default.
- W4253313247 modified "2023-10-02" @default.
- W4253313247 title "ChemPix: Automated Recognition of Hand-drawn Hydrocarbon Structures Using Deep Learning" @default.
- W4253313247 doi "https://doi.org/10.26434/chemrxiv.14156957" @default.
- W4253313247 hasPublicationYear "2021" @default.
- W4253313247 type Work @default.
- W4253313247 citedByCount "1" @default.
- W4253313247 countsByYear W42533132472023 @default.
- W4253313247 crossrefType "posted-content" @default.
- W4253313247 hasAuthorship W4253313247A5003577701 @default.
- W4253313247 hasAuthorship W4253313247A5040462364 @default.
- W4253313247 hasAuthorship W4253313247A5057261669 @default.
- W4253313247 hasAuthorship W4253313247A5057461346 @default.
- W4253313247 hasAuthorship W4253313247A5062125510 @default.
- W4253313247 hasAuthorship W4253313247A5064472074 @default.
- W4253313247 hasBestOaLocation W42533132471 @default.
- W4253313247 hasConcept C108583219 @default.
- W4253313247 hasConcept C115961682 @default.
- W4253313247 hasConcept C119857082 @default.
- W4253313247 hasConcept C153180895 @default.
- W4253313247 hasConcept C154945302 @default.
- W4253313247 hasConcept C157657479 @default.
- W4253313247 hasConcept C199360897 @default.
- W4253313247 hasConcept C2777904410 @default.
- W4253313247 hasConcept C41008148 @default.
- W4253313247 hasConcept C50644808 @default.
- W4253313247 hasConcept C81363708 @default.
- W4253313247 hasConceptScore W4253313247C108583219 @default.
- W4253313247 hasConceptScore W4253313247C115961682 @default.
- W4253313247 hasConceptScore W4253313247C119857082 @default.
- W4253313247 hasConceptScore W4253313247C153180895 @default.
- W4253313247 hasConceptScore W4253313247C154945302 @default.
- W4253313247 hasConceptScore W4253313247C157657479 @default.
- W4253313247 hasConceptScore W4253313247C199360897 @default.
- W4253313247 hasConceptScore W4253313247C2777904410 @default.
- W4253313247 hasConceptScore W4253313247C41008148 @default.
- W4253313247 hasConceptScore W4253313247C50644808 @default.
- W4253313247 hasConceptScore W4253313247C81363708 @default.
- W4253313247 hasLocation W42533132471 @default.
- W4253313247 hasLocation W42533132472 @default.
- W4253313247 hasOpenAccess W4253313247 @default.
- W4253313247 hasPrimaryLocation W42533132471 @default.
- W4253313247 hasRelatedWork W2731899572 @default.
- W4253313247 hasRelatedWork W2999805992 @default.
- W4253313247 hasRelatedWork W3116150086 @default.
- W4253313247 hasRelatedWork W3133861977 @default.
- W4253313247 hasRelatedWork W4200173597 @default.
- W4253313247 hasRelatedWork W4223943233 @default.
- W4253313247 hasRelatedWork W4291897433 @default.
- W4253313247 hasRelatedWork W4312417841 @default.
- W4253313247 hasRelatedWork W4321369474 @default.
- W4253313247 hasRelatedWork W4380075502 @default.
- W4253313247 isParatext "false" @default.
- W4253313247 isRetracted "false" @default.
- W4253313247 workType "article" @default.