Text Mining And Analysis Of Harbin Ice And Snow World Scenic Spot Comments

In recent years,with the gradual development of transportation,tourist travel has become a normal holiday.At the same time,with the rapid development of network technology,people’s demand for network platform is also increasing.Comment on scenic spots,services and other aspects through the network platform;The users’ comments on their opinions and attitudes will also have a corresponding impact on the choice of scenic spots by follow-up tourists and,to some extent,affect the economic situation of local tourism,especially under the impact of the epidemic,the income of tourism in many regions has been falling year after year.Therefore,how to efficiently identify,extract and analyze the emotions and needs of tourists from the comments of scenic spots,so as to provide consumers with more effective services.The main purpose of this paper is to promote the development of local tourism economy while enhancing its attraction to tourists.This paper mainly applies text mining technology to sentiment analysis of comments on the famous scenic spots of Harbin Ice and Snow World,and studies the comment data on ctrip,Dianping and other platforms.Firstly,the emotional words in the commentary text are classified by the Dalian Institute of Technology Emotional Dictionary,and the keywords under different emotions are summarized and analyzed.For some unincluded emotional words,the method of searching for synonyms is adopted to expand the dictionary,so as to improve the emotional recognition degree of the dictionary to the commentary text.According to the expanded sentiment dictionary,the comments were scored for their emotional polarity,and the comments with the highest and lowest scores were analyzed.According to the polarity of emotions combined with manual judgment,positive and negative emotions are labeled for the comment text,and emotion classification is carried out for the annotated comment text through the classifier model commonly used in machine learning.The accuracy of the comment text under different classifiers is compared,and the model with the highest accuracy is selected to analyze the classification results.Finally,the LDA method is used to model the theme of the favorable and unfavorable comments of the review text,summarize the theme under different topics,and sort out the tourist preference,so as to provide more targeted improvement schemes for the decision makers of scenic spots.According to the analysis results,the scenic area can improve the safety protection devices of slides,increase the number of scenic tour performances,set up a small range of artificial snowfall sites,and set up the park strategy slides to strengthen the strengths of the park.At the same time,the negative evaluation of tourists can be reduced by adding special taxi channels,providing hot water points and reducing excessive publicity.In this way,the pain points of tourists can be reduced and the purpose of attracting tourists can be achieved.At the same time,tourists should also read relevant comments before visiting and make adequate preparations to better visit the scenic spot.The model helps managers to design activities of scenic spots according to the needs and preferences of tourists and enrich the contents of scenic spots.For tourists,they can also define their own concerns and preferences and needs according to this,to prevent passenger miscarriage.