Extracting chemical food safety hazards from the scientific literature automatically using large language models

The number of scientific articles published in the domain of food safety has consistently been increasing over the last few decades. It has therefore become unfeasible for food safety experts to read all relevant literature related to food safety and the occurrence of hazards in the food chain. Howe...

Full description

Saved in:
Bibliographic Details
Main Authors: Neris Özen, Wenjuan Mu, Esther D. van Asselt, Leonieke M. van den Bulk
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Applied Food Research
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2772502224002890
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850102114962898944
author Neris Özen
Wenjuan Mu
Esther D. van Asselt
Leonieke M. van den Bulk
author_facet Neris Özen
Wenjuan Mu
Esther D. van Asselt
Leonieke M. van den Bulk
author_sort Neris Özen
collection DOAJ
description The number of scientific articles published in the domain of food safety has consistently been increasing over the last few decades. It has therefore become unfeasible for food safety experts to read all relevant literature related to food safety and the occurrence of hazards in the food chain. However, it is important that food safety experts are aware of the newest findings and can access this information in an easy and concise way. In this study, an approach is presented to automate the extraction of chemical hazards from the scientific literature through large language models. The large language model was used out-of-the-box and applied on scientific abstracts; no extra training of the models or a large computing cluster was required. Three different styles of prompting the model were tested to assess which was the most optimal for the task at hand. The prompts were optimized with two validation foods (leafy greens and shellfish) and the final performance of the best prompt was evaluated using three test foods (dairy, maize and salmon). The specific wording of the prompt was found to have a considerable effect on the results. A prompt breaking the task down into smaller steps performed best overall. This prompt reached an average accuracy of 93 % and contained many chemical contaminants already included in food monitoring programs, validating the successful retrieval of relevant hazards for the food safety domain. The results showcase how valuable large language models can be for the task of automatic information extraction from the scientific literature.
format Article
id doaj-art-463df6847df2461ebfb0d89afd0af901
institution DOAJ
issn 2772-5022
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series Applied Food Research
spelling doaj-art-463df6847df2461ebfb0d89afd0af9012025-08-20T02:39:50ZengElsevierApplied Food Research2772-50222025-06-015110067910.1016/j.afres.2024.100679Extracting chemical food safety hazards from the scientific literature automatically using large language modelsNeris Özen0Wenjuan Mu1Esther D. van Asselt2Leonieke M. van den Bulk3Wageningen Food Safety Research, Wageningen University & Research, Wageningen, The NetherlandsWageningen Food Safety Research, Wageningen University & Research, Wageningen, The NetherlandsWageningen Food Safety Research, Wageningen University & Research, Wageningen, The NetherlandsCorresponding author.; Wageningen Food Safety Research, Wageningen University & Research, Wageningen, The NetherlandsThe number of scientific articles published in the domain of food safety has consistently been increasing over the last few decades. It has therefore become unfeasible for food safety experts to read all relevant literature related to food safety and the occurrence of hazards in the food chain. However, it is important that food safety experts are aware of the newest findings and can access this information in an easy and concise way. In this study, an approach is presented to automate the extraction of chemical hazards from the scientific literature through large language models. The large language model was used out-of-the-box and applied on scientific abstracts; no extra training of the models or a large computing cluster was required. Three different styles of prompting the model were tested to assess which was the most optimal for the task at hand. The prompts were optimized with two validation foods (leafy greens and shellfish) and the final performance of the best prompt was evaluated using three test foods (dairy, maize and salmon). The specific wording of the prompt was found to have a considerable effect on the results. A prompt breaking the task down into smaller steps performed best overall. This prompt reached an average accuracy of 93 % and contained many chemical contaminants already included in food monitoring programs, validating the successful retrieval of relevant hazards for the food safety domain. The results showcase how valuable large language models can be for the task of automatic information extraction from the scientific literature.http://www.sciencedirect.com/science/article/pii/S2772502224002890Chemical contaminationFood safetyInformation extractionPrompt engineeringNatural language processingArtificial intelligence
spellingShingle Neris Özen
Wenjuan Mu
Esther D. van Asselt
Leonieke M. van den Bulk
Extracting chemical food safety hazards from the scientific literature automatically using large language models
Applied Food Research
Chemical contamination
Food safety
Information extraction
Prompt engineering
Natural language processing
Artificial intelligence
title Extracting chemical food safety hazards from the scientific literature automatically using large language models
title_full Extracting chemical food safety hazards from the scientific literature automatically using large language models
title_fullStr Extracting chemical food safety hazards from the scientific literature automatically using large language models
title_full_unstemmed Extracting chemical food safety hazards from the scientific literature automatically using large language models
title_short Extracting chemical food safety hazards from the scientific literature automatically using large language models
title_sort extracting chemical food safety hazards from the scientific literature automatically using large language models
topic Chemical contamination
Food safety
Information extraction
Prompt engineering
Natural language processing
Artificial intelligence
url http://www.sciencedirect.com/science/article/pii/S2772502224002890
work_keys_str_mv AT nerisozen extractingchemicalfoodsafetyhazardsfromthescientificliteratureautomaticallyusinglargelanguagemodels
AT wenjuanmu extractingchemicalfoodsafetyhazardsfromthescientificliteratureautomaticallyusinglargelanguagemodels
AT estherdvanasselt extractingchemicalfoodsafetyhazardsfromthescientificliteratureautomaticallyusinglargelanguagemodels
AT leoniekemvandenbulk extractingchemicalfoodsafetyhazardsfromthescientificliteratureautomaticallyusinglargelanguagemodels