Research On Influencing Factors Of Mobile Phone Customer Satisfaction Based On Online Shopping Reviews

Nowadays society,with the continuous development of internet technology,more and more people choose to buy what they need through online shopping.With the continuous increase of transaction volume,the amount of online review data is also increasing.Online review data generally includes information related to business services,text narration closely related to commodity experience,and commodity pictures,videos and other information.These information directly show people's most real and direct feelings of goods.And these feelings constitute the influencing factors of customer satisfaction.Therefore,it is very important to mine the information related to users' use feelings from online review data to obtain the influencing factors,and then study the influence degree of these influencing factors on customer satisfaction.This paper first introduces the background and significance of the research on satisfaction,and through the analysis and summary of the research status at home and abroad,we have a general understanding of the research on satisfaction.Then we use crawler technology to crawl the comments of iPhone 12 mini,Mate 40 Pro,OPPO Reno5 Pro and vivo X60 in Jingdong Mall,and obtain the specific data of the comments.Firstly,through browsing to view the specific content of the data,the invalid or repeated comments are filtered and deleted to ensure the availabili-ty of the comment data.Secondly,the high-frequency words in the comments are extracted and processed.Finally,the 10 characteristic factors affecting satisfaction are determined as follows:appearance,color,performance,price,logistics,photog-raphy,battery,screen,sound quality and physical properties.Finally,the five point scoring standard is adopted to score the data according to Likert quantitative ta-ble,and the data set for establishing satisfaction model is obtained.In order to show the superiority of the model,the model is compared with the model based on neural network and random forest.The accuracy,precision and recall of the model are selected as the evaluation indexes of the model.It is found that the customer satisfaction model based on Bayesian has the best measurement effect.The cus-tomer satisfaction model based on Bayesian can output the importance ranking of the influencing factors of customer satisfaction of each mobile phone product and the relationship between the various factors.According to the degree of influence of each factor on customer satisfaction,we can put forward relevant opinions and suggestions to manufacturers.Attributes that have a high impact on user satisfaction,such as oppo reno5 pro's physical attributes and photo attributes,have a great impact on customer satisfaction.Manufacturers should give full play to the advantages of their prod-ucts and turn the advantages of physical attributes and photo attributes into sales focus,so as to improve the sales quota.For attributes that have a low impact,for example,the photo attribute of vivo X60 is not very prominent,and the cus-tomer's satisfaction with this function is general.Manufacturers need to strengthen the technical research on this attribute,and put forward corresponding solutions.Therefore,studying the influencing factors of mobile phone customer satisfaction is conducive to the manufacturers to improve the performance of their products and make timely strategic adjustments to meet the market demand and improve their core competitiveness.