Sensing The Spatio-temporal Variations Of Tourist Sentiment And Topic Mining Research

The sentiment and distribution of visitors is a direct reflection of the quality of scenic area development and an important reference for tourists in choosing a scenic area.Understanding the distribution characteristics,emotional changes and review themes of tourists can provide decision support for scenic area development planning,as well as a reference for tourists to choose a scenic area.Online reviews are an important medium for understanding changes in visitor sentiment,and an important player in characterising visitor distribution.With the popularisation of the Internet and the accelerated development of big data analytics,especially the deepening of natural language processing technologies such as sentiment analysis and theme mining bring new opportunities for sensing visitor distribution,sentiment change and commentary research.This paper takes Chinese 5A scenic area tourists as an example for empirical analysis,and uses tourists’ comments on online tourism platforms as the main data source.It tries to combine the traditional theoretical and practical research content of tourists’ evaluation with the spatio-temporal analysis method based on emotion classification,and constructs a more complete and effective research method system for perceiving tourists’ emotional spatio-temporal changes and comment themes,which is mainly divided into two major aspects: tourists’ emotional spatio-temporal distribution pattern The system is divided into two major aspects: visitor sentiment spatio-temporal distribution pattern mining and theme mining,in which visitor sentiment spatio-temporal distribution pattern mining mainly uses structured statistical data to explore the changing characteristics of visitor sentiment in different spatiotemporal areas from a quantitative perspective;theme mining uses unstructured text to explore the main comments made by visitors while playing from a qualitative perspective.The following main conclusions were drawn.(1)The number of visitors to China’s 5A scenic spots is negatively correlated with visitor sentiment: the number of visitors is highest in October and lower in March;the value of visitor sentiment is highest in March and lowest in October.(2)The temporal and spatial distribution of tourists has obvious aggregation:temporally mainly in July,August and October,spatially mainly in the Yangtze River Delta city cluster,Beijing-Tianjin-Hebei city cluster and Guanzhong Plain city cluster.(3)There are significant differences in the emotional hot and cold spots of tourists in each city group: the Yangtze River Delta city group is always the emotional hot spot;the northern city group has more emotional cold spots;the central city group changes significantly within the year;the southwestern city group has more emotional hot spots.(4)Both positive and negative comments from tourists focus on the aspect of service attitude,and negative comments mainly focus on ticketing,service and transportation,etc.In-depth mining of negative comments from tourists is an important way to improve the development quality of scenic spots.The research system of perceiving the spatio-temporal characteristics of tourists’ emotions built on the basis of natural language processing methods and geospatial analysis methods in this paper can deepen the understanding of emotional geography in the tourism industry,further explore the spatio-temporal differences in tourists’ emotions as well as their experiences and demands in the process of tourism,and provide some thoughts and inspiration to the decision-making of scenic spots and the high-quality development of regional tourism.