Research On Huawei Mobile Phone Review Based On Data Mining Technology

2019 is a year of explosive growth of Internet users.According to the 44 th statistical report on the development of China's Internet,China's Internet users have reached 854 million and mobile Internet users have reached 847 million.In this era when the Internet has entered thousands of households,e-commerce platforms are also popular among people.The Internet is an information sharing platform,each major online shopping platform has a huge amount of consumer purchase review data.These data have gradually become the basis for consumers to make purchase decisions and the important reference for manufacturers to upgrade products.However,how to mine valuable and referential information from these vast text review data has become a new challenge.The purpose of this paper is to use the current mainstream data mining technology to investigate the review data of Huawei mate30 mobile phones,and analyze the characteristics of mobile phones that consumers pay attention to at present,so as to provide a reference for mobile phone manufacturers to optimize and upgrade their products and enhance their market competitiveness as well as for consumers to make purchase decisions.Firstly,this paper adopts the crawler technology of Python software based on Ubuntu system to collect the data of user comments on Huawei mate30 smartphone on "Jingdong Mall" and "Taobao" platform,then uses Excel to standardize the data,and then uses the current popular Jieba software to segment the cleaned data and mark the part of speech.Then we use the word cloud technology of wordle software visualization to show the features in the data.Finally,according to the segmentation data,we build LDA model and use k-means clustering to analyze the emotional orientation to obtain the emotional characteristics of consumers' attention.The final research results show that using open source system tools Linux Ubuntu and R language,based on the consumer comment information including JD and Taobao,using the current popular LDA model and K-means clustering method for user emotion recognition has more accurate analysis results than single analysis comment method.Data analysis shows that this method can not only accurately analyze the emotional tendency of the research object,but also infer the characteristics of the smartphone that consumers most need for goods.The final result shows that the operation speed,charging and other performance of Huawei's mate30 mobile phone are satisfactory to consumers,while the mobile phone heating,after-sales service and gifts are not in line with consumers' expectations.From the analysis results of Jingdong Mall,we can see that the consumers of the platform pay more attention to the performance of goods and distribution services;and based on the data analysis results of Taobao,we can see that the consumers of the platform pay more attention to the whole process experience of consumption,such as after-sales and gift service of mobile phones.Therefore,the results of this study can be used as a reference for e-commerce platform to optimize sales services,consumers' decision-making behavior when purchasing mobile phones,and manufacturers' timely improvement and upgrading of their own products.