Web Online Review Text Mining

With the rapid development of e-commerce industry,as well as the construction and continuous improvement of e-commerce website service platform,a large number of online consumer reviews related to goods or services will appear every day.It contains rich content information,which has great use value for businesses or consumers who choose goods or services.It is an important reference for online consumers to choose products or services,and businesses to improve their own products and services.However,in the face of a large number of online review text data,it is unrealistic to use the method of manual reading to obtain valuable information in comments.Therefore,this paper takes the tourism online review data as an example,based on the Chinese text mining,uses R software and ROST CM6 software to mine and analyze the tourism online evaluation data,in order to quickly and accurately extract the key information with utilization value,so as to improve the utilization value of the online review text data.At the same time,the research work of this paper also provides a reference idea for the mining analysis of online review text data.The Chinese text mining method used in this paper mainly includes semantic network analysis,topic model and emotional tendency analysis.First of all,through the relativity measurement of high-frequency words and semantic network analysis,this paper makes a preliminary mining and analysis of the text data,and extracts the key information in the comment text data;Secondly,through the modeling and analysis of the topic model,it realizes the extraction of the topic words in the comment text data and the classification of the topic of the comment,so as to obtain the different categories in the comment content more intuitively and quickly;then,the emotional classification of comment text data is realized by using the method of emotional tendentiousness analysis,and the comments of emotional polar errors are identified and analyzed;Finally,the paper combines the emotional tendentiousness analysis with the topic model,and through the topic modeling of the results of emotional tendentiousness analysis,the tourists'emotional tendentiousness judgment of various services in the scenic spot is realized.In addition,this paper also combines the results of text mining and analysis with the actual situation,and puts forward some reference suggestions for tourists to choose tourist attractions,make travel plans,and also for scenic spots to improve their own facilities and services,and develop accurate marketing service strategies.