2021
DOI: 10.2196/28229
|View full text |Cite
|
Sign up to set email alerts
|

A Fine-Tuned Bidirectional Encoder Representations From Transformers Model for Food Named-Entity Recognition: Algorithm Development and Validation

Abstract: Background Recently, food science has been garnering a lot of attention. There are many open research questions on food interactions, as one of the main environmental factors, with other health-related entities such as diseases, treatments, and drugs. In the last 2 decades, a large amount of work has been done in natural language processing and machine learning to enable biomedical information extraction. However, machine learning in food science domains remains inadequately resourced, which brings… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
5
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6
2
1
1

Relationship

3
7

Authors

Journals

citations
Cited by 19 publications
(8 citation statements)
references
References 43 publications
0
5
0
Order By: Relevance
“…Providing an annotated corpus with regard to different semantic resources allows us to further train information extraction methods using the state of the art in NLP. These resources were further utilized for training NER methods and their utility has already been published in [ 23 ].…”
Section: Discussionmentioning
confidence: 99%
“…Providing an annotated corpus with regard to different semantic resources allows us to further train information extraction methods using the state of the art in NLP. These resources were further utilized for training NER methods and their utility has already been published in [ 23 ].…”
Section: Discussionmentioning
confidence: 99%
“…In the specific domain of food computing, a series of research works have applied state-of-the-art models on unique food-centric corpora. For instance, explorations leveraging the FoodBase dataset (Popovski et al, 2019) have utilized models like Bidirectional LSTM (BiLSTM) (Cenikj et al, 2020), and fine-tuning BERT and its variants (Perera et al, 2022;Stojanov et al, 2021;Cenikj et al, 2021). Further advances in the field have been made following the presentation of the TASTEset dataset (Wróblewska et al, 2022), introducing a novel benchmark through the strategic fine-tuning of BERT integrated with a CRF layer.…”
Section: Named Entity Recognitionmentioning
confidence: 99%
“…The second corpus-based NER method for the food domain, FoodNER [20], performs finetuning of BERT [21] and BioBERT [9] for the NER task using the annotations from the FoodBase corpus. Apart from identifying the mentions of food entities from the text, the FoodNER model is also capable of performing NEL, i.e.…”
Section: Corpus-based Named Entity Recognitionmentioning
confidence: 99%