BioPREP: Deep learning-based predicate classification with SemMedDB

Hong, Gi-Bong; Kim, Yuheun; Choi, Yeon-Jung; Song, Min

doi:10.1016/j.jbi.2021.103888

Cited by 7 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Selected Studies Precision, recall, F1 score [112], [38], [37], [56], [54], [53], [39], [28], [59], [77], [65], [66], [47], [36], [49], [74], [50], [22], [59], [52], [55], [21], [35], [113], [32], [68], [62], [41], [40], [60], [33], [81], [57], [76], [46], [71], [44], [48] Alignment with gold standard [114], [112], [22], [60], [76], [38]…”

Section: Evaluation Metricsmentioning

confidence: 99%

A Systematic Review on Semantic Role Labeling for Information Extraction in Low-Resource Data

Ariyanto,

Purwitasari,

Fatichah

2024

IEEE Access

View full text Add to dashboard Cite

Challenges in the big data phenomenon arise due to the existence of unstructured text data, which is very large, comes from various sources, has various formats, and contains much noise. The complexity of unstructured text data makes it difficult to extract useful information. Therefore, a process is needed to transform it into structured data to be processed further. The information Extraction (IE) process helps to extract relationships, entities, semantic roles, and events from unstructured text data by converting them into structured output. One of IE's tasks is Semantic Role Labeling (SRL), which has a crucial function in identifying semantic roles in a sentence so that it can enrich the understanding of the text. However, much of SRL development focuses on high-resource data, especially in English. The limited development of SRL in specific low-resource languages or domains is a complex challenge. This research aims to conduct a systematic study on the development of SRL for low-resource data, both in low-resource language or domainspecific contexts. The review process was carried out systematically using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) model, and 54 quality papers were obtained from the filtering process (from 2018 to 2023). We review several essential points, including (1) datasets that are often used for SRL tasks and their labeling strategies for low-resource data, (2) methods that have currently been developed for SRL tasks and learning scenarios when dealing with low-resource data, (4) evaluation metrics, (5) application of SRL tasks. This review is complemented by a discussion of issues and potential solutions for developing SRL on low-resource data to help researchers develop SRL more effectively in dealing with the challenges faced with low-resource data.

show abstract

Section: Evaluation Metricsmentioning

confidence: 99%

A Systematic Review on Semantic Role Labeling for Information Extraction in Low-Resource Data

Ariyanto,

Purwitasari,

Fatichah

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…With the implementation of the BioBERT model [43][44][45][46], Natural Language Processing tasks extract better relations and generate more accurate outcomes. Instead of pre-training on generic data sets, BioBert requires derived data sets to perform well.…”

Section: Biobert Modelmentioning

confidence: 99%

A hybrid algorithm for clinical decision support in precision medicine based on machine learning

Zhang

Lin

2023

BMC Bioinformatics

View full text Add to dashboard Cite

Purpose The objective of the manuscript is to propose a hybrid algorithm combining the improved BM25 algorithm, k-means clustering, and BioBert model to better determine biomedical articles utilizing the PubMed database so, the number of retrieved biomedical articles whose content contains much similar information regarding a query of a specific disease could grow larger. Design/methodology/approach In the paper, a two-stage information retrieval method is proposed to conduct an improved Text-Rank algorithm. The first stage consists of employing the improved BM25 algorithm to assign scores to biomedical articles in the database and identify the 1000 publications with the highest scores. The second stage is composed of employing a method called a cluster-based abstract extraction to reduce the number of article abstracts to match the input constraints of the BioBert model, and then the BioBert-based document similarity matching method is utilized to obtain the most similar search outcomes between the document and the retrieved morphemes. To realize reproducibility, the written code is made available on https://github.com/zzc1991/TREC_Precision_Medicine_Track. Findings The experimental study is conducted based on the data sets of TREC2017 and TREC2018 to train the proposed model and the data of TREC2019 is used as a validation set confirming the effectiveness and practicability of the proposed algorithm that would be implemented for clinical decision support in precision medicine with a generalizability feature. Originality/value This research integrates multiple machine learning and text processing methods to devise a hybrid method applicable to domains of specific medical literature retrieval. The proposed algorithm provides a 3% increase of P@10 than that of the state-of-the-art algorithm in TREC 2019.

show abstract

“…The current leading NLP models such as BERT [20], GPT [21], and T5 [22] announced later are all based on this transformer block. In particular, BERT is commonly used in biomedical text mining research because it is built on multiple transformers encoder blocks, which has the advantage of compressing the sentence and mining semantic information from it [8,[23][24][25].…”

Section: Deep Learning-based Semantic Relation Classification Modelmentioning

confidence: 99%

“…Hong et al [ 25 ] created a dataset labeled with predicate relations by performing automatic NER on SemMedDB data and then clustering on predicates that appear with the recognized entity pairs. Various deep learning models were trained with the dataset to verify the usability of their dataset and propose the final state-of-the-art performance model optimized with the dataset.…”

Section: Introductionmentioning

confidence: 99%

BertSRC: transformer-based semantic relation classification

Lee

Son

Song

2022

BMC Med Inform Decis Mak

Self Cite

View full text Add to dashboard Cite

The relationship between biomedical entities is complex, and many of them have not yet been identified. For many biomedical research areas including drug discovery, it is of paramount importance to identify the relationships that have already been established through a comprehensive literature survey. However, manually searching through literature is difficult as the amount of biomedical publications continues to increase. Therefore, the relation classification task, which automatically mines meaningful relations from the literature, is spotlighted in the field of biomedical text mining. By applying relation classification techniques to the accumulated biomedical literature, existing semantic relations between biomedical entities that can help to infer previously unknown relationships are efficiently grasped. To develop semantic relation classification models, which is a type of supervised machine learning, it is essential to construct a training dataset that is manually annotated by biomedical experts with semantic relations among biomedical entities. Any advanced model must be trained on a dataset with reliable quality and meaningful scale to be deployed in the real world and can assist biologists in their research. In addition, as the number of such public datasets increases, the performance of machine learning algorithms can be accurately revealed and compared by using those datasets as a benchmark for model development and improvement. In this paper, we aim to build such a dataset. Along with that, to validate the usability of the dataset as training data for relation classification models and to improve the performance of the relation extraction task, we built a relation classification model based on Bidirectional Encoder Representations from Transformers (BERT) trained on our dataset, applying our newly proposed fine-tuning methodology. In experiments comparing performance among several models based on different deep learning algorithms, our model with the proposed fine-tuning methodology showed the best performance. The experimental results show that the constructed training dataset is an important information resource for the development and evaluation of semantic relation extraction models. Furthermore, relation extraction performance can be improved by integrating our proposed fine-tuning methodology. Therefore, this can lead to the promotion of future text mining research in the biomedical field.

show abstract

BioPREP: Deep learning-based predicate classification with SemMedDB

Cited by 7 publications

References 22 publications

A Systematic Review on Semantic Role Labeling for Information Extraction in Low-Resource Data

A Systematic Review on Semantic Role Labeling for Information Extraction in Low-Resource Data

A hybrid algorithm for clinical decision support in precision medicine based on machine learning

BertSRC: transformer-based semantic relation classification

Contact Info

Product

Resources

About