Spatial-temporal Changes And Theme Exploration Of Tourists' Emotional Experience Based On Weibo Data

The problem of the world is the problem of human being.Emotion is an important indicator of the distinction between man and the world.It is the essential way of perceiving the past,present and future.Emotion research is not only the focus of sociology and psychology,but also an important research direction of geography.At certain times and in specific places.people will produce various emotions such as pain,joy,anger and so on.Incorporating the emotional experience into the study of geography will help explain the diversity of human behavior and social life.As a special group,tourists visit different places to seek spiritual satisfaction and aesthetic pleasure,and in the process,different emotions are generated due to various factors.Exploring the generation,spatio-temporal variation and influences of these emotions is an important part of the study of "emotional turn,in tourism and geography.The widespread use of social media,geo-tagged social media data provides the possibility to study the spatio-temporal variation of the emotional experience of tourists.Taking Xi'an as a case study,and the tourists' emotional experience as the research object,this study collects the micro-blog data of domestic tourists of three years,and updates the existing sentiment analysis model based on Weibo data to calculate the emotional value.This study explores the spatio-temporal variation of domestic tourists' emotional experience in Xi'an during the three years from the two dimensions of positive emotion and negative emotion.The hot spot grid map method and Getis-Ord Git method are used to explore the spatio-temporal variation of tourists9 emotional experience and hot-cold spot gathering.Based on these analyses,the X-means clustering method is used to analyze the variation of emotional experience for three years,and finally the regular pattern of tourists' emotional experience spatio-temporal variation is obtained.Finally,this study uses LDA topic model to explore the topic of microblog texts of tourists under different emotional states,refines and summarizes the microblog emotions related to tourism,and extracts tourism related feature words for auxiliary verification.Main conclusions as follow:(1)From a spatial point of view,tourists in the main tourism function areas have a high positive emotional experience.The high-emotional experience area of Xi'an is mainly the Xi'an Ancient City Tourism Area and the Qujiang New District.The transportation hub(station,high-speed entrance and exit)is mainly a gathering place of negative emotional experience,while the emotional experience of tourists at the edge of the city is low;from the perspective of time,the overall change of tourists' emotions is not obvious in three years.The sentiment difference between adjacent two years is mainly concentrated in the range of[-10,10].(2)From a spatio-temporal point of view,tourists in the city center,around Chang'an South Road,major commercial districts and scenic spots are relatively high and stable,with agglomeration effect.During the past three years,the positive emotional distribution of tourists gradually spread from the center of the city and scenic spots to the surrounding areas,and the distribution of cold spots gradually began to spread from the edge of the city and the import and export of transportation.The negative emotions of tourists were scattered around the cold spots.And the cold spots of negative emotions were concentrated on the edge of the city and the import and export areas of urban traffic.(3)From the results of spatio-temporal clustering,the overall change of tourists,emotions in Xi'an has little fluctuation in the past three years.The temporal and spatial clustering results of positive emotions and negative emotions have certain similarities.They all show three types of changes:the stable,relatively stable and drastic fluctuations.The stable type is mainly located in the center of the city,the main business districts,the traffic trunk line and the scenic spots;the relatively stable in most areas of Xi'an;the drastic fluctuations are mainly located in the edge of the city.(4)Using the LDA topic model,we explored 11 tourists9 positive emotional topics including local diet,scenic spots,infrastructure,urban landscape,urban local sense,cultural heritage,tourist companions,tourism goods,weather,festival activities,and travel itinerary;11 tourists' negative emotional topics are also found,including local diet,service quality,transportation facilities,cultural heritage,physical condition,urban management,public security environment,travel itinerary,escape from real life,weather,and travel order.Local diet,infrastructure,cultural heritage,travel itinerary and weather have an impact both on positive and negative emotional experiences.In the past three years,there has been no significant change in the attention of tourists in various aspects of Xi'an.Local food,infrastructure,scenic spots and tourist companions have always been the focus of tourists,and tourists are paying more attention to the emotional experience brought about by interpersonal interaction today.(5)Through the topic excavation of the micro-blog text in tourist's emotional cold and hot spots,there are 8 tourism-related topics in tourist positive emotion hot spots(urban landscape,local diet,infrastructure,weather,scenic spots,festival activities,special festivals and service quality)and 4 topics in tourist positive emotion cold spots(transportation facilities,festival activities,escape from real life and weather).Tourists,negative emotion hot spots lead to 2 tourism-related topics(public security environment and infrastructure).Besides,7 tourism-related topics are also found in tourists' negative emotion cold spots,which are namely infrastructure,escape from real life,weather?local diet,service quality,public security environment and physical condition.The results of the topic exploration are basically in line with the characteristics of tourists,hot and cold spots,which explains to some extent the reasons for the distribution of tourist emotion spatial distribution.The innovations of this research mainly include the followings:(1)Explore The spatio-temporal variation and regular pattern of tourists' emotional experience.This study visualizes the tourists1 emotions temporal and spatial distribution.and finds that the high-emotional experience area of Xi'an are mainly the Xi'an Ancient City Tourism Area and the Qujiang New District,and the low-emotional experience area are the transportation hub and the edge of the city.Meanwhile,tourists in the city center?around Chang'an South Road,major commercial districts and scenic spots are relatively high and stable.(2)Explore the theme of Weibo text based on tourists' emotions.This study introduces the LDA theme model to explore the theme of tourists' positive emotions and negative emotions microblog texts,and finds that there are 11 topics,such as local diet,scenic spots,infrastructure affecting tourists' positive emotional experience.Besides,there are 11 topics,such as local diet,service quality,transportation facilities affecting tourists' negative emotional experience.