城市旅游餐饮体验的注意力机制模型建构——基于机器学习的网络文本深度挖掘-《南京师大学报（自然科学版）》

文章信息/Info

Title:: Construction of Attention Mechanism Model of Urban Tourism Catering Experience:Deep Mining of Online Text Based on Machine Learning

作者:: 张旭辉¹; 2; 张郴¹; 2; 李雅南¹; 2; 徐梓榆¹; 2; 黄震方¹; 2; (1.南京师范大学地理科学学院,江苏南京 210023)(2.江苏省地理信息资源开发与利用协同创新中心,江苏南京 210023)

Author(s):: Zhang Xuhui¹; 2; Zhang Chen¹; 2; Li Yanan¹; 2; Xu Ziyu¹; 2; Huang Zhenfang¹; 2; (1.School of Geography,Nanjing Normal University,Nanjing 210023,China)(2.Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application,Nanjing 210023,China)

摘要:: 良好的旅游餐饮体验对提升城市旅游竞争力的意义不可小觑. 旅游餐饮文本记录了游客对旅游目的地餐饮店的产品和服务的真实体验,文本内容海量且受时空约束较小,颇具一定的分析价值. 本文将Attention机制(注意力机制)融合机器学习方法中的卷积神经网络模型,旨在对餐饮文本背后的深层语义关联进行深度挖掘,探寻影响游客积极餐饮体验的激励因素和消极餐饮体验的保健因素及其影响机制,这适用于旅游餐饮文本自身特质. 通过注意力机制模型建构,发现影响游客对南京餐饮体验评价的保健因素由价格、区位交通和服务组成,而激励因素则由口味、名气和文化环境组成. 另外,本文对注意力权重排序,优先解决权重较高因素,提升游客餐饮体验.

Abstract:: The significance of a good tourist catering experience in enhancing the competitiveness of urban tourism cannot be underestimated. The tourist catering text records the real experience of tourists to the products and services of the catering shops in the tourist destination. The text has a large amount of content and is less constrained by time and space,which has certain analytical value. This paper integrates Attention mechanism with the convolutional neural network model in the machine learning method,and aims to deeply explore the deep semantic associations behind the catering text,and explore the motivational factors that affect tourists’ positive catering experience and the healthcare factors of negative catering experience and their influencing mechanisms,which are applicable to the characteristics of tourist catering texts themselves. Through the construction of attention mechanism model,it is found that the healthcare factors that affect tourists’ uation of Nanjing’s catering experience are composed of price,location,transportation and services,while the motivational factors are composed of taste,fame and cultural environment. In addition,this paper sorts the value of attention weight,and prioritizes the resolution of higher-weight factors to improve the catering experience of tourists.