The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models
In order to promote the digital dissemination and preservation of Chinese intangible cultural heritage, this work constructs a digital platform for its transmission. The platform integrates a range of advanced technologies, including the Densely Connected Convolutional Networks - Bottleneck and Comp...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S240584402417017X |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841526292506214400 |
|---|---|
| author | Zhurong Liu |
| author_facet | Zhurong Liu |
| author_sort | Zhurong Liu |
| collection | DOAJ |
| description | In order to promote the digital dissemination and preservation of Chinese intangible cultural heritage, this work constructs a digital platform for its transmission. The platform integrates a range of advanced technologies, including the Densely Connected Convolutional Networks - Bottleneck and Compression model, a notable convolutional neural network, along with natural language processing algorithms, generative adversarial network algorithms, and neural collaborative filtering algorithms. The platform is validated with 224,055 publicly archived valid data records, ensuring its effectiveness and reliability. The results indicate that the heteroscedasticity-robust standard error is 0.15, and the platform achieves accuracies of 0.87 and 0.89 in 5-fold and 10-fold cross-validation, respectively, demonstrating the model's stability and predictive accuracy. Under various operating conditions, the platform's accuracy, recall, and F1-Score all exceed 0.85, with a root mean square error below 0.12, ensuring efficient and reliable recommendation performance. Regarding user experience, page load times remain steady between 6.45 and 10.25 s, with bounce rates and error rates controlled between 40.88 % to 58.33 % and 0.05 %–0.18 %, respectively. A/B testing and heatmap analysis indicate that optimizing page layout significantly improves click-through and conversion rates. Additionally, in the fields of oral literature and traditional medicine, the platform achieves entity recognition accuracies of 90.2 % and 91.3 %, respectively. The platform also demonstrates strong coverage in the knowledge of folklore and traditional crafts, with rates reaching 92.5 % and 89.8 %, respectively, showcasing its advantages across different domains. This work aims to integrate cultural heritage conservation, cultural inheritance, and digital dissemination, fostering the broad transmission and preservation of intangible cultural heritage in the era of big data. |
| format | Article |
| id | doaj-art-a88aee0d068f4a7f9b99b7e7586a5e3a |
| institution | Kabale University |
| issn | 2405-8440 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj-art-a88aee0d068f4a7f9b99b7e7586a5e3a2025-01-17T04:49:57ZengElsevierHeliyon2405-84402025-01-01111e40986The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network modelsZhurong Liu0School of International Tourism and Culture, Guizhou Normal University, Guiyang, 550025, ChinaIn order to promote the digital dissemination and preservation of Chinese intangible cultural heritage, this work constructs a digital platform for its transmission. The platform integrates a range of advanced technologies, including the Densely Connected Convolutional Networks - Bottleneck and Compression model, a notable convolutional neural network, along with natural language processing algorithms, generative adversarial network algorithms, and neural collaborative filtering algorithms. The platform is validated with 224,055 publicly archived valid data records, ensuring its effectiveness and reliability. The results indicate that the heteroscedasticity-robust standard error is 0.15, and the platform achieves accuracies of 0.87 and 0.89 in 5-fold and 10-fold cross-validation, respectively, demonstrating the model's stability and predictive accuracy. Under various operating conditions, the platform's accuracy, recall, and F1-Score all exceed 0.85, with a root mean square error below 0.12, ensuring efficient and reliable recommendation performance. Regarding user experience, page load times remain steady between 6.45 and 10.25 s, with bounce rates and error rates controlled between 40.88 % to 58.33 % and 0.05 %–0.18 %, respectively. A/B testing and heatmap analysis indicate that optimizing page layout significantly improves click-through and conversion rates. Additionally, in the fields of oral literature and traditional medicine, the platform achieves entity recognition accuracies of 90.2 % and 91.3 %, respectively. The platform also demonstrates strong coverage in the knowledge of folklore and traditional crafts, with rates reaching 92.5 % and 89.8 %, respectively, showcasing its advantages across different domains. This work aims to integrate cultural heritage conservation, cultural inheritance, and digital dissemination, fostering the broad transmission and preservation of intangible cultural heritage in the era of big data.http://www.sciencedirect.com/science/article/pii/S240584402417017XDensely connected convolutional networksBottleneck and compression modelConvolutional neural networkIntangible cultural heritageDigital dissemination platformCultural inheritance |
| spellingShingle | Zhurong Liu The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models Heliyon Densely connected convolutional networks Bottleneck and compression model Convolutional neural network Intangible cultural heritage Digital dissemination platform Cultural inheritance |
| title | The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models |
| title_full | The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models |
| title_fullStr | The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models |
| title_full_unstemmed | The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models |
| title_short | The construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models |
| title_sort | construction of a digital dissemination platform for the intangible cultural heritage using convolutional neural network models |
| topic | Densely connected convolutional networks Bottleneck and compression model Convolutional neural network Intangible cultural heritage Digital dissemination platform Cultural inheritance |
| url | http://www.sciencedirect.com/science/article/pii/S240584402417017X |
| work_keys_str_mv | AT zhurongliu theconstructionofadigitaldisseminationplatformfortheintangibleculturalheritageusingconvolutionalneuralnetworkmodels AT zhurongliu constructionofadigitaldisseminationplatformfortheintangibleculturalheritageusingconvolutionalneuralnetworkmodels |