Worldwide city transport typology prediction with sentence-BERT based supervised learning via Wikipedia

Rath, Srushti; Chow, Joseph Y.J.

doi:10.1016/j.trc.2022.103661

Cited by 8 publications

(3 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These findings suggest that CoSENT may be better suited for capturing semantic similarity between sentences [ 79 ], particularly when more powerful pre-trained models are employed. However, it is noteworthy that all combinations of models yielded Pearson's and Spearman's coefficients above 0.75, indicating that pre-trained language models and sentence embedding schemes are effective for capturing semantic similarity between sentences, corroborating the view of studies such as [ 66 – 68 ].…”

Section: Resultssupporting

confidence: 74%

“…Such approaches require significant computational resources and struggle to make real-time inferences, which can be alleviated by the representation-based twin-tower structure with sentence embedding at its core [ 66 ]. Sentence-BERT [ 67 ] obtains sentence embedding by siamese and triplet network structure and the semantic similarity of two sentences can be gauged by the cosine similarity between their embeddings [ 68 ]. A Simple contrastive learning framework of sentence embedding (SimCSE), including both unsupervised and supervised versions, was proposed by Gao et al in [ 69 ], achieving the SOTA performance of sentence-level semantic representation based on contrast learning and dropout data augment [ 70 ].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Plant disease prescription recommendation based on electronic medical records and sentence embedding retrieval

Ding,

Qiao,

Zhang

2023

Plant Methods

View full text Add to dashboard Cite

Background In the era of Agri 4.0 and the popularity of Plantwise systems, the availability of Plant Electronic Medical Records has provided opportunities to extract valuable disease information and treatment knowledge. However, developing an effective prescription recommendation method based on these records presents unique challenges, such as inadequate labeling data, lack of structural and linguistic specifications, incorporation of new prescriptions, and consideration of multiple factors in practical situations. Results This study proposes a plant disease prescription recommendation method called PRSER, which is based on sentence embedding retrieval. The semantic matching model is created using a pre-trained language model and a sentence embedding method with contrast learning ideas, and the constructed prescription reference database is retrieved for optimal prescription recommendations. A multi-vegetable disease dataset and a multi-fruit disease dataset are constructed to compare three pre-trained language models, four pooling types, and two loss functions. The PRSER model achieves the best semantic matching performance by combining MacBERT, CoSENT, and CLS pooling, resulting in a Pearson coefficient of 86.34% and a Spearman coefficient of 77.67%. The prescription recommendation capability of the model is also verified. PRSER performs well in closed-set testing with Top-1/Top-3/Top-5 accuracy of 88.20%/96.07%/97.70%; and slightly worse in open-set testing with Top-1/Top-3/Top-5 accuracy of 82.04%/91.50%/94.90%. Finally, a plant disease prescription recommendation system for mobile terminals is constructed and its generalization ability with incomplete inputs is verified. When only symptom information is available without environment and plant information, our model shows slightly lower accuracy with Top-1/Top-3/Top-5 accuracy of 75.24%/88.35%/91.99% in closed-set testing and Top-1/Top-3/Top-5 accuracy of 75.08%/87.54%/89.84% in open-set testing. Conclusions The experiments validate the effectiveness and generalization ability of the proposed approach for recommending plant disease prescriptions. This research has significant potential to facilitate the implementation of artificial intelligence in plant disease treatment, addressing the needs of farmers and advancing scientific plant disease management.

show abstract

Section: Resultssupporting

confidence: 74%

Section: Related Workmentioning

confidence: 99%

Plant disease prescription recommendation based on electronic medical records and sentence embedding retrieval

Ding,

Qiao,

Zhang

2023

Plant Methods

View full text Add to dashboard Cite

show abstract

“…Initially, many doubted the quality of Wikipedia's content and allegations of misinformation, so several institutions prohibited teachers and students from using Wikipedia [7], [9], [10]. However, this assumption is gradually changing because Wikipedia is supported by millions of editors and moderators of multi-sourced content [11], [12]. This research focuses on whether the digital library Wikipedia usage in biology learning contributes to improving students' scientific literacy to support biology learning with innovative digital media that has not been used previously.…”

Section: Introductionmentioning

confidence: 99%

Developing Students’ Scientific Literacy through Wikipedia Digital Library

Toharudin,

Damaianti,

Hizqiyah

et al. 2023

Int. J. Emerg. Technol. Learn.

View full text Add to dashboard Cite

This study investigates the impact of utilizing the digital library Wikipedia on high school students’ scientific literacy skills in biology. This research used a quasi-experiment design with a one-group pre-test and post-test method, conducted in a high school in Bandung, Indonesia. The participants consisted of 60 tenth-grade students. The research instrument consisted of 24 multiple-choice questions designed to assess nine scientific literacy indicators, including the ability to describe phenomena scientifically, design and evaluate scientific investigations, and interpret data and evidence scientifically. Each question was scored dichotomously, with the correct answer receiving a score of 1 and incorrect answers assigned a score of 0. Pre-test and post-test scores were analyzed using the Rasch model through Winsteps Version 3.73. Based on data analysis, the items fit with the Rasch model because the infit and outfit MNSQ is 1.0 and the residual item of unexplained variance is 2.3% to 9.8%. The reliability of the items shows an item value of 0.97, which is classified as excellent. The indicators considered the most difficult before being treated were identifying accurate scientific ideas, distinguishing between types of sources, and evaluating literature source validity after the treatment. There was a significant improvement in students’ scientific literacy skills after utilizing the digital library, Wikipedia. The findings provide valuable insights into the effectiveness of integrating digital media into biology courses, highlighting the potential of Wikipedia as a digital reference source for enhancing students’ scientific literacy in high school settings.

show abstract

Enabling CMF estimation in data-constrained scenarios: A semantic-encoding knowledge mining model

Qi,

Li,

Zhang

2024

Accident Analysis & Prevention

View full text Add to dashboard Cite

Worldwide city transport typology prediction with sentence-BERT based supervised learning via Wikipedia

Cited by 8 publications

References 39 publications

Plant disease prescription recommendation based on electronic medical records and sentence embedding retrieval

Plant disease prescription recommendation based on electronic medical records and sentence embedding retrieval

Developing Students’ Scientific Literacy through Wikipedia Digital Library

Enabling CMF estimation in data-constrained scenarios: A semantic-encoding knowledge mining model

Contact Info

Product

Resources

About