Transfer Learning for Information Extraction with Limited Data

Nguyen, Minh-Tien; Phan, Viet-Anh; Linh, Le Thai; Sơn, Nguyễn Hồng; Dung, Le; Hirano, Miku; Hotta, Hajime

doi:10.1007/978-981-15-6168-9_38

Cited by 9 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It usually uses fine-tuning that adapts pretrained ML models with target training samples (Gururangan et al, 2020). In business scenarios, TL can help to reduce the number of training samples (Nguyen et al, 2020;.…”

Section: Seq2seq and Attentionmentioning

confidence: 99%

Extend Learning Space With the Smartphone for Students at Hung Yen University of Technology and Education, Vietnam.

Nguyen¹,

Nguyen²

2019

View full text Add to dashboard Cite

The advent of smartphones as a communication revolution, it is just a portable multimedia device that everybody uses it. A smartphone can help you a lot in work and entertainment such as phone call, e-mail, video record, touch-and-go payment, navigation, interpretation, special useful in surfing the Internet. Students can use smartphones as a powerful learning mean. However, many universities in Vietnam do not allow smartphone usage in the classroom, because students may be distracted by smartphone applications and loose focusing on learning. In fact, students have been not instructed how to use smartphones effectively in the classroom. Recently, in the context of technological development and globalization, students need to study in an open space and open education. It means that teaching and learning are not just restricted classroom with chalk and talk, but teachers and students have more degree of freedom in terms of space, time and means to achieve learning outcomes. Students are free to choose the means, methods, and styles of study. The proposed research aims to expand the learning space for students at Hung Yen University of Technical Education by smartphone and focus on solving the following issues:1) Can smartphones be used as means of learning or not? 2) How to effectively use the smartphone in the classroom? (Student guidance use the smartphone in their study, how to control student's behaviors, classroom arrangement, outcome assessment, student response). To conduct this research, we designed lessons based on mobile applications. Pilot on 02 students groups in the 3rd year at Hung Yen University of Technology and Education to assess the impact of the organization of teaching with smart phones to the excitement, the active learning and the response of students and teachers. We used the method of classroom observation, surveys, direct interviews with students and lecturers on the test results. From there, propose recommendations on lesson design and organize teaching with smart phones in Hung Yen University of Technology and Education.

show abstract

Section: Seq2seq and Attentionmentioning

confidence: 99%

Extend Learning Space With the Smartphone for Students at Hung Yen University of Technology and Education, Vietnam.

Nguyen¹,

Nguyen²

2019

View full text Add to dashboard Cite

show abstract

“…However, it has the human annotation of three NLP tasks which are beneficial for the analysis of incident reports. In addition, a small number of samples is still helpful in business scenarios for training AI models by using transfer learning (Devlin et al, 2019;Nguyen et al, 2020.…”

Section: Quantitative Observationmentioning

confidence: 99%

Towards Safer Operations: An Expert-involved Dataset of High-Pressure Gas Incidents for Preventing Future Failures

Inoue,

Nguyen,

Mizokuchi

et al. 2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: Industry Track

View full text Add to dashboard Cite

This paper introduces a new IncidentAI dataset for safety prevention. Different from prior corpora that usually contain a single task, our dataset comprises three tasks: named entity recognition, cause-effect extraction, and information retrieval. The dataset is annotated by domain experts who have at least six years of practical experience as high-pressure gas conservation managers. We validate the contribution of the dataset in the scenario of safety prevention. Preliminary results on the three tasks show that NLP techniques are beneficial for analyzing incident reports to prevent future failures. The dataset facilitates future research in NLP and incident management communities. The access to the dataset is also provided. 1

show abstract

“…We employed a pre-trained BERT as the embedding layer due to two reasons: (i) BERT has achieved promising results on 11 NLP tasks, including sequence tagging [5,13] and (ii) BERT was contextually trained on a huge amount of data, so it can encode word meaning, which is important for machine learning models.…”

Section: Embedding Layermentioning

confidence: 99%

Sentence Compression as Deletion with Contextual Embeddings

Nguyen¹,

Minh²,

Le³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Sentence compression is the task of creating a shorter version of an input sentence while keeping important information. In this paper, we extend the task of compression by deletion with the use of contextual embeddings. Different from prior work usually using non-contextual embeddings (Glove or Word2Vec), we exploit contextual embeddings that enable our model capturing the context of inputs. More precisely, we utilize contextual embeddings stacked by bidirectional Long-short Term Memory and Conditional Random Fields for dealing with sequence labeling. Experimental results on a benchmark Google dataset show that by utilizing contextual embeddings, our model achieves a new state-of-theart F-score compared to strong methods reported on the leader board.

show abstract

Transfer Learning for Information Extraction with Limited Data

Cited by 9 publications

References 12 publications

Extend Learning Space With the Smartphone for Students at Hung Yen University of Technology and Education, Vietnam.

Extend Learning Space With the Smartphone for Students at Hung Yen University of Technology and Education, Vietnam.

Towards Safer Operations: An Expert-involved Dataset of High-Pressure Gas Incidents for Preventing Future Failures

Sentence Compression as Deletion with Contextual Embeddings

Contact Info

Product

Resources

About