游客对空气质量感知的情感分析——基于双向LSTM模型  

Sentiment Analysis of Tourists′Air Quality Perception——Based on Bidirectional LSTM Model

作　　者：刘文佳 陶玉国 LIU Wen-jia;TAO Yu-guo(College of History and Culture and Tourism,Jiangsu Normal University,Xuzhou 221116,China)

机构地区：[1]江苏师范大学历史文化与旅游学院,江苏徐州221116

出　　处：《资源开发与市场》2020年第9期929-935,共7页Resource Development & Market

基　　金：国家自然科学基金项目“区域旅游业碳排放的测度、影响因素与碳减排效应研究——以江苏省为例”(编号:41571131);江苏省社会科学基金项目“江苏旅游业碳排放测度及其因素分解”(编号:I5GLB015);江苏师范大学研究生创新项目“基于图片和文字UGC的中国国内游客队旅游目的地气候变化情感分析”(编号:2019XKT242);江苏省高校创新训练项目“基于社交媒体数据的中国游客对区域空气质量的情感分析”(编号:201910320099Z)。

摘　　要：空气质量是影响旅游目的地发展的重要因素,游客目的地选择、旅游体验和重游意愿等与其关系紧密。获取2013年1月至2019年7月游客在携程网发布的259个国内5A级景区空气质量的13,069条评论,通过双向LSTM模型进行情感分析,借助Gephi、Vosviewer分析其共现网络图,并依托扎根理论诊断影响游客对景区空气质量感知的因素。结果发现:①游客对我国旅游目的地空气质量感知情感值正面极性显著,时间上呈稳中有升的趋势,空间上“南高北低”格局明显。②高情感值景区与景区类型相关,低情感值景区则受地理位置影响。③共现网络图谱和扎根理论分析均得出自然环境、身体健康和服务设施是影响情感值的主要因素。Analysis of tourists′perception of air quality in tourist destinations was of great significance to development of tourism destina-tions.Obtaining 13,069 comments published by C-trip about air quality of 2595A-level scenic spots in China from January 1,2013 to July 31,2019.Using the bidirectional LSTM model to calculate the evaluation scores,and analyzed co-occurrence network with the help of Gephi、Vosviewer,diagnosed the influencing factors of perception with grounded theory.The results showed that:①Tourists′perception of the senti-ment value of air quality in Chinese tourist destinations had a significant positive polarity,and showed a steadily increasing trend over time,and the spatial pattern of“south high and low north”was obvious.②High emotional value scenic spot was closely related totourist attraction types,while scenic spots with low emotional values were greatly affected by geographic location.③The results of co-occurrence network and ground-ed theory indicated that the influencing factors was constituted by natural environment,physicalhealth and service facilities.

关 键 词：空气质量 情感分析 双向LSTM模型 

分 类 号：F713.55[经济管理—市场营销] F590-05[经济管理—产业经济]