Lv Qi
School of Tourism Culture, The Tourism College of Changchun University, Changchun, 130607, China.
Sci Rep. 2025 Apr 26;15(1):14698. doi: 10.1038/s41598-025-99635-z.
This paper investigates the participation of industry-school cooperation and talent cultivation models in tourism schools within the framework of the internet of things (IoT) by employing a back propagation (BP) prediction model. The paper delves into the participation model of industry-school cooperation under the IoT context, elucidating the application and optimization of the BP prediction model in the specific environment of tourism education. Experimental results validate the proposed model's effectiveness, demonstrating substantial advantages over traditional models. Compared to support vector machine (SVM) and random forest (RF) models, the optimized BP model exhibits marked improvements in accuracy, precision, mean squared error, and prediction time. In tests conducted on various data types, the optimized model achieves an accuracy rate of 87.3% for textual data, 88.7% for numerical data, and 86.2% for categorical data, significantly outperforming both SVM and RF models. Regarding prediction time, the optimized model processes 1,000 data entries in just 0.05 s, which is considerably faster than SVM (0.12 s) and RF (0.08 s). Moreover, the model demonstrates superior performance in mean squared error, achieving errors of 0.063 for textual data, 0.059 for numerical data, and 0.067 for categorical data, consistently surpassing the baseline models across all metrics. Based on the predictions generated by the BP model, this paper recommends strengthening collaborations between educational institutions and enterprises, expanding the scope of partnership projects, and increasing internship opportunities for students. These findings provide valuable guidance for optimizing school-enterprise collaboration models and talent cultivation strategies within tourism education. Such improvements are anticipated to enhance students' practical skills and employability while fostering innovation in the development of tourism talent. This paper also expands the interdisciplinary applications of the BP prediction model within the educational sector, offering new perspectives and methodologies for enhancing school-enterprise collaborations in tourism education. By advancing these frameworks, the paper contributes to the innovation and evolution of educational models, promoting both practical advancements and theoretical progress in the field.
本文通过采用反向传播（BP）预测模型，研究了物联网（IoT）框架下旅游院校的产教融合与人才培养模式。本文深入探讨了物联网背景下的产教融合参与模式，阐明了BP预测模型在旅游教育特定环境中的应用与优化。实验结果验证了所提模型的有效性，表明其相较于传统模型具有显著优势。与支持向量机（SVM）和随机森林（RF）模型相比，优化后的BP模型在准确率、精确率、均方误差和预测时间方面均有显著提升。在对各种数据类型进行的测试中，优化后的模型对文本数据的准确率达到87.3%，对数值数据的准确率为88.7%，对分类数据的准确率为86.2%，明显优于SVM和RF模型。在预测时间方面，优化后的模型处理1000条数据记录仅需0.05秒，比SVM（0.12秒）和RF（0.08秒）快得多。此外，该模型在均方误差方面表现出色，文本数据的误差为0.063，数值数据的误差为0.059，分类数据的误差为0.067，在所有指标上均持续超过基线模型。基于BP模型生成的预测结果，本文建议加强教育机构与企业之间的合作，扩大合作项目范围，并增加学生的实习机会。这些研究结果为优化旅游教育中的校企合作模式和人才培养策略提供了有价值的指导。预计这些改进将提高学生的实践技能和就业能力，同时促进旅游人才培养的创新。本文还拓展了BP预测模型在教育领域的跨学科应用，为加强旅游教育中的校企合作提供了新的视角和方法。通过推进这些框架，本文为教育模式的创新与发展做出了贡献，推动了该领域的实践进步和理论发展。
