Scholarlink

<p>To address the issues of insufficient utilization of multimodal data and limited model robustness in the prediction methods for tourism brand influence, firstly, this study adopts a Transformer-based pre-trained model as the feature extractor to generate embedded representations of multimodal data, including text, images, and audience feedback. Secondly, an attention mechanism is introduced through a multimodal attention-aware module to achieve in-depth fusion of cross-modal features, thereby enhancing the model’s understanding of the correlation between film and television communication and tourism brands. Thirdly, the Adaboost ensemble learning algorithm and the Catch Fish Optimization Algorithm (CFOA) are combined to improve the stability and accuracy of prediction results. Finally, based on the effect of film and television communication, the influence of tourism brands is predicted. The research results show that the prediction accuracy of the proposed model reaches 94.1%, 91.3%, and 93.7% on text data, audience feedback data, and image data respectively, and 92.5% on multimodal fusion data, all of which are superior to other models. The results of ablation experiments indicate that the attention mechanism, Adaboost, and CFOA all make significant contributions to the model performance, among which the attention mechanism is the most critical for cross-modal information fusion. In addition, in the context of film and television communication, the proposed model achieves the highest prediction accuracy for each indicator of Anhui Province’s tourism brand, which is an improvement compared with the context without film and television communication. This fully verifies the role of film and television communication in enhancing the predictability of brand influence. This study aims to improve the prediction performance of tourism brand influence and also provides intelligent decision support for the construction of Anhui’s tourism brand. Furthermore, it promotes the continuous improvement of brand value of the regional tourism industry in the environment of digital communication.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The Adaboost and CFOA algorithms for implementation of prediction model in Anhui tourism brand based on multimodal data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月6日

查看全文
  • Tingting Wang,
  • Panpan Chang

摘要

To address the issues of insufficient utilization of multimodal data and limited model robustness in the prediction methods for tourism brand influence, firstly, this study adopts a Transformer-based pre-trained model as the feature extractor to generate embedded representations of multimodal data, including text, images, and audience feedback. Secondly, an attention mechanism is introduced through a multimodal attention-aware module to achieve in-depth fusion of cross-modal features, thereby enhancing the model’s understanding of the correlation between film and television communication and tourism brands. Thirdly, the Adaboost ensemble learning algorithm and the Catch Fish Optimization Algorithm (CFOA) are combined to improve the stability and accuracy of prediction results. Finally, based on the effect of film and television communication, the influence of tourism brands is predicted. The research results show that the prediction accuracy of the proposed model reaches 94.1%, 91.3%, and 93.7% on text data, audience feedback data, and image data respectively, and 92.5% on multimodal fusion data, all of which are superior to other models. The results of ablation experiments indicate that the attention mechanism, Adaboost, and CFOA all make significant contributions to the model performance, among which the attention mechanism is the most critical for cross-modal information fusion. In addition, in the context of film and television communication, the proposed model achieves the highest prediction accuracy for each indicator of Anhui Province’s tourism brand, which is an improvement compared with the context without film and television communication. This fully verifies the role of film and television communication in enhancing the predictability of brand influence. This study aims to improve the prediction performance of tourism brand influence and also provides intelligent decision support for the construction of Anhui’s tourism brand. Furthermore, it promotes the continuous improvement of brand value of the regional tourism industry in the environment of digital communication.