Application Of Text Mining In User Reviews Of Online Shopping

In 2020,we have already stepped into a new era-big data era.Informatization is the general trend of the development of the times.In daily life,people are creating a lot of data every day,and these data often contain a lot of unexplored information.Therefore,by means of data integration and data mining technology,these information can be found and play its value,as a reference to provide data support for decision-making,and provide convenience for people's lives.Under the background of this era,online shopping has already entered thousands of households,and more and more people choose the convenient Internet to buy all kinds of goods needed for daily life.Naturally,there is also a large number of online shopping user comment data,it is very difficult to extract valuable information from such large-scale text data,and text mining technology can bring hope to solve this problem.This article uses Python crawler to crawl the user comments data of Huawei and apple mobile on Taobao and Jd.com e-commerce platforms,through LDA theme modeling and sentiment analysis,determine the focus and demand of mobile phone users,and how satisfied they are with the features of each phone.Through this research,on the one hand,it can help users to further understand the advantages and disadvantages of each mobile phone,provide reference information for them,and then assist in purchase decision-making.On the other hand,it can help merchants fully understand users' demands and concerns for mobile phones,as well as the existing problems and advantages of various types of mobile phones,so as to clarify the next optimization and improvement direction and enhance the user satisfaction of products.The research of this paper is mainly carried out from four parts.The first part is data acquisition,which uses Python crawler tool to crawl the user comment data of Huawei mobile phone and apple mobile phone on two online shopping platforms,Taobao(including tmall mall)and Jingdong Mall.The second part mainly deals with data cleaning,dealing with redundant data,improving data quality,modifying data format,and segmenting comment data with Jieba segmentation.The third part is based on the visual analysis of word cloud,and the establishment of LDA topic model for text semantic mining.The fourth part uses Baidu API to analyze the sentiment tendency of Huawei and apple mobile users' comments based on the results of LDA theme model.The final analysis shows that,Mobile users focus on features such as function,configuration,performance and price,In addition,users also pay attention to the shopping experience of the shopping process.According to the comparison of Huawei and apple's sentiment analysis,Huawei's mobile phone is superior in three aspects:photo effect,endurance and cost-effective,while Apple's phone is superior in sound quality,sound effect and system operation.