Research On Customer's Satisfaction Index Based On Unstructured Text From The Perspective Of Big Data

In this era of rapid development of the Internet,e-commerce enterprises are highly competitive.Customer satisfaction is related to the survival of enterprises,and it affects the sustainable development of e-commerce in a country or region.Carrying out customer satisfaction research,we can understand and master the consumer awareness of the brand,loyalty and satisfaction,and provide reference for the relevant industry enterprises to enhance the brand value,which is of great significance to promote the implementation of national brand strategy,and to cultivate a number of brands with international influence.This paper mainly uses the method of big data analysis to study customer satisfaction based on unstructured text data.The main contents of this paper include the following four aspects.Firstly,according to the influence factors of customer satisfaction,this paper preliminarily sets up the CSI evaluation model suitable for this study,based on the full study of the domestic and foreign satisfaction model and the theory of structural equation model.Secondly,through the python web crawler,this paper captures the product reviews of seven electrical digital brand products on the four major shopping sites(Tmall,Jingdong Mall,Gome Online,and Suning).And then,this paper finishes the participle for the product reviews using the method of jieba,constructs the consumer satisfaction review thesaurus combined with the commentary semantics,and abstracts the entity vocabulary that affects the consumer's satisfaction with the brand from the commentary and quantifies it with the modifier.Finally,the structured text data is applied to the previously proposed CSI evaluation model,and the model is improved to improve the CSI evaluation model,which is suitable for the text data.The customer satisfaction index is calculated by estimating the model parameters,and using the method of performance analysis,this paper summarizes the ways to enhance the brand satisfaction in the specific entity level.There are two main innovations in this paper.Firstly,under the framework of the existing CSI,this paper transforms the traditional subjective data collection method,which is based on the questionnaire survey,to collect the objective and abundant consumer's commodity review data,which is more objective and more real to reflect the views of consumers on the purchased goods.Secondly,in this paper,the text mining from the perspective of big data is introduced into the traditional satisfaction measure model,which provides a new idea for the study of satisfaction model.Through the research,this paper argues that the key to enhance the brand needs to start from the three aspects of perceived quality,expected quality and brand image.Specific to the entity level,to enhance the brand needs to start from the appearance,feel,shell,call,battery and other product quality,and logistics,customer service and other services,and promotions,discounts,discounts and other promotional aspects and reputation,publicity,integrity and other aspects of brand image.