<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In this paper, a CNN-Bi-LSTM-CRF model is formed by using convolutional neural network combined with Bi-LSTMCRF, and at the same time, with the help of the attention mechanism, which is used to carry out attribute extraction research on the attribute-valued texts in the intangible cultural heritage annotated intangible cultural heritage attribute corpus as well as the validation of the automatic text detection system for the error text, and the results show that the CNN-Bi-LSTM-CRF model performs optimally in the attribute extraction of “heritage type”, with a score of 97.382. The results show that the CNN-Bi-LSTM-CRF model has the best performance, with a score of 97.382 in the attribute extraction of “heritage type”, and the average detection error rate of the traditional system is (17.5%,14.20%) in the comparison of the detection error rate of handwritten and printed text, respectively, the average detection error rate of Bi-LSTM-CRF is (7.19%, 6.15%). The average detection error rate is (7.19%, 6.15%). CNN-Bi-LSTM-CRF samples and categorizes the translations of the three NRLs, and finds that there is a considerable proportion of mistranslations in the linguistic dimension (30.2%), cultural dimension (8.05%) and communicative dimension (10.07%). Finally, it explores tourists’ satisfaction with the destination through outreach factors. Most foreign tourists’ affective polarity scores for the Great Wall range from 0.1 to 0.3, with primarily positive emotions.</p>
</abstract>

In this paper, a CNN-Bi-LSTM-CRF model is formed by using convolutional neural network combined with Bi-LSTMCRF, and at the same time, with the help of the attention mechanism, which is used to carry out attribute extraction research on the attribute-valued texts in the intangible cultural heritage annotated intangible cultural heritage attribute corpus as well as the validation of the automatic text detection system for the error text, and the results show that the CNN-Bi-LSTM-CRF model performs optimally in the attribute extraction of “heritage type”, with a score of 97.382. The results show that the CNN-Bi-LSTM-CRF model has the best performance, with a score of 97.382 in the attribute extraction of “heritage type”, and the average detection error rate of the traditional system is (17.5%,14.20%) in the comparison of the detection error rate of handwritten and printed text, respectively, the average detection error rate of Bi-LSTM-CRF is (7.19%, 6.15%). The average detection error rate is (7.19%, 6.15%). CNN-Bi-LSTM-CRF samples and categorizes the translations of the three NRLs, and finds that there is a considerable proportion of mistranslations in the linguistic dimension (30.2%), cultural dimension (8.05%) and communicative dimension (10.07%). Finally, it explores tourists’ satisfaction with the destination through outreach factors. Most foreign tourists’ affective polarity scores for the Great Wall range from 0.1 to 0.3, with primarily positive emotions.

Research on Intangible Cultural Heritage Outreach Translation Based on Natural Language Processing Technology

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In this paper, a CNN-Bi-LSTM-CRF model is formed by using convolutional neural network combined with Bi-LSTMCRF, and at the same time, with the help of the...

Research on Intangible Cultural Heritage Outreach Translation Based on Natural Language Processing Technology

Online veröffentlicht: 26. Feb. 2024

Seitenbereich: -

Eingereicht: 15. Jan. 2024

Akzeptiert: 20. Jan. 2024

DOI: https://doi.org/10.2478/amns-2024-0618

Schlüsselwörter
Attention mechanism, Natural language processing, Automatic detection, Bi-LSTM-CRF modeling

© 2023 Nan Dong, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Research on Intangible Cultural Heritage Outreach Translation Based on Natural Language Processing Technology

Online veröffentlicht: 26. Feb. 2024

Seitenbereich: -

Eingereicht: 15. Jan. 2024

Akzeptiert: 20. Jan. 2024

DOI: https://doi.org/10.2478/amns-2024-0618

SchlüsselwörterAttention mechanism, Natural language processing, Automatic detection, Bi-LSTM-CRF modeling

© 2023 Nan Dong, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Schlüsselwörter
Attention mechanism, Natural language processing, Automatic detection, Bi-LSTM-CRF modeling