Data-Augmented Hybrid Named Entity Recognition for Disaster Management by Transfer Learning

Kung, Hung-Kai; Hsieh, Chun-Mo; Ho, Chia-Cheng; Tsai, Yun-Cheng; Chan, Hao-Yung; Tsai, Meng-Han

doi:10.3390/app10124234

Cited by 10 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To overcome the limitation, a system for damage collection and report process is required. In addition, techniques for unstructured report processing [28,41,42] may be adopted to enable automatic analysis and further reduce the labor and time cost.…”

Section: Discussionmentioning

confidence: 99%

“…In Taiwan, Tsai et al [27] proposed a three-module conversation-based system framework, implemented as a chatbot to notify and support school building managers completing damage inspections and report submissions after earthquakes. Kung et al [28] adopted transfer learning to automatically extract information from the collected damage reports and provide analysis for decision-makers to understand the situation.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A Chatbot System to Support Mine Safety Procedures during Natural Disasters

Tsai

Chan

et al. 2021

Sustainability

Self Cite

View full text Add to dashboard Cite

This study developed a chatbot to improve the efficiency of government activation of mine safety procedures during natural disasters. Taiwan has a comprehensive governmental system dedicated to responding to frequent natural disasters, and the Bureau of Mines has instituted clear procedures to ensure the delivery of disaster alarms and damage reports. However, the labor- and time-consumption procedures are inefficient. In this study, we propose a system framework for disaster-related information retrieval and immediate notifications to support the execution of mine safety procedures. The framework utilizes instant messaging (IM) applications as the user interface to look up information and send messages to announce the occurrence of disaster events. We evaluated the efficiency of the procedures before and after adopting the system and achieved a time-cost reduction of 55.8 min among three types of disaster events. The study has proven the feasibility of adopting novel techniques for decision-making and assures the improvement of the efficiency and effectiveness of the procedure activation.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

A Chatbot System to Support Mine Safety Procedures during Natural Disasters

Tsai

Chan

et al. 2021

Sustainability

Self Cite

View full text Add to dashboard Cite

show abstract

“…Furthermore, an approach using a combination of active learning and self-learning was proposed to reduce the workload required for the NER task involving tweet streams [43]. Certainly, transfer learning has also been widely used in NER; for example, a mandarin NER module based on a transfer learning system was constructed for collecting and analyzing disaster information in disaster management [44]. To address the lack of labeled data in artifact recognition, [45] proposed a combination of BiLSTM and a CRF for named artifact entity recognition, which is a semisupervised model that uses labeled data to conduct training to achieve efective recognition performance.…”

Section: Ner In Aviationmentioning

confidence: 99%

A TL_FLAT Model for Chinese Text Datasets of UAV Power Systems: Optimization and Performance

Shen

Yang

et al. 2023

International Journal of Intelligent Systems

View full text Add to dashboard Cite

The manufacturing processes of unmanned aerial vehicle (UAV) power systems generate large amounts of data and knowledge. The extraction of useful information or patterns from redundant data and knowledge texts has become a challenge in intelligent manufacturing. Unfortunately, graphics processing unit (GPU)-based parallel computing is limited, and the inference speeds of the available named entity recognition (NER) models for Chinese text datasets are low because they are mainly based on the long short-term memory (LSTM) algorithm. Herein, first, the flat-lattice transformer (FLAT) model was optimized by using a stochastic gradient descent with momentum (SGDM) optimizer and adjusting the model hyperparameters. Compared with the existing NER methods, the proposed optimization algorithm achieved better performance on the available dataset. Then, an NER method named the TL_FLAT model based on transfer learning and the abovementioned optimization model was introduced. Finally, a Chinese text dataset from a UAV power system created by the authors was used to validate the proposed method. The F1 score was 76.26%, the precision value was 76.98%, and the recall value was 75.56%, indicating that the TL_FLAT model was suitable for Chinese text entity recognition for UAV power systems.

show abstract

“…The CRFs are also chosen as an algorithm for selecting highly reliable cases. Kung [35] built a mandarin NER module-based transfer learning system to do damage information collection and analysis in disaster management. To deal with the cultural relic areas' lack of labeled data, Zhang et al [36] put forward a model, Cultural Relics SCRNER (semi-supervision model for Cultural Relics' Named Entity Recognition), using tag data no training BiLSTM and CRF model to obtain the effective identification performance.…”

Section: Text Data Augmented In Nermentioning

confidence: 99%

Fine-Grained Mechanical Chinese Named Entity Recognition Based on ALBERT-AttBiLSTM-CRF and Transfer Learning

et al. 2020

View full text Add to dashboard Cite

Manufacturing text often exists as unlabeled data; the entity is fine-grained and the extraction is difficult. The above problems mean that the manufacturing industry knowledge utilization rate is low. This paper proposes a novel Chinese fine-grained NER (named entity recognition) method based on symmetry lightweight deep multinetwork collaboration (ALBERT-AttBiLSTM-CRF) and model transfer considering active learning (MTAL) to research fine-grained named entity recognition of a few labeled Chinese textual data types. The method is divided into two stages. In the first stage, the ALBERT-AttBiLSTM-CRF was applied for verification in the CLUENER2020 dataset (Public dataset) to get a pretrained model; the experiments show that the model obtains an F1 score of 0.8962, which is better than the best baseline algorithm, an improvement of 9.2%. In the second stage, the pretrained model was transferred into the Manufacturing-NER dataset (our dataset), and we used the active learning strategy to optimize the model effect. The final F1 result of Manufacturing-NER was 0.8931 after the model transfer (it was higher than 0.8576 before the model transfer); so, this method represents an improvement of 3.55%. Our method effectively transfers the existing knowledge from public source data to scientific target data, solving the problem of named entity recognition with scarce labeled domain data, and proves its effectiveness.

show abstract

Data-Augmented Hybrid Named Entity Recognition for Disaster Management by Transfer Learning

Cited by 10 publications

References 17 publications

A Chatbot System to Support Mine Safety Procedures during Natural Disasters

A Chatbot System to Support Mine Safety Procedures during Natural Disasters

A TL_FLAT Model for Chinese Text Datasets of UAV Power Systems: Optimization and Performance

Fine-Grained Mechanical Chinese Named Entity Recognition Based on ALBERT-AttBiLSTM-CRF and Transfer Learning

Contact Info

Product

Resources

About