Research On Quality Evaluation And Improvement Of Electronic Products Based On Text Mining And QFD

With the development of the Internet and the modern logistics industry,people are more inclined to use mobile payment methods for online shopping.Online shopping has also grown in size,and online reviews provide consumers with an ample opportunity to describe their shopping experience and evaluate the combined attributes of products,services,and manufacturers.Users express their emotional tendencies and the key factors that affect their emotions through comments,The higher the quality of the comments,the more information they can help manufacturers and users learn more about goods and services,and guide reasonable product quality improvements in the supply chain.decision making.Therefore,how to mine high-value and high-efficiency review texts based on customer satisfaction,and how to extract valuable text feature method systems are the main focus of attention.Quality Function Deployment(QFD)is a quality management tool for companies today to enhance users’ satisfaction.The key to effectively using this tool is to seek real users’ demand information,The traditional way to obtain demand information is to conduct questionnaire surveys,but its results are relatively subjective,and the observable data sources and data volumes are also small.The online reviews of laptops of different brands are used as the data source to mine actual needs of users through data analysis,and improve the traditional QFD model.First,draw the text word cloud map,network semantic map and related word frequency table for the 6,000 pieces of data obtained by the web crawler,and conduct customer demand analysis based on the Textrank algorithm and LDA topic model,and extract and mine the key empirical knowledge in online reviews.And build a customer satisfaction index system for electronic products,and analyze and calculate the index weight through the CCLSTM deep learning model.Secondly,after determining the customer demand index,the QFD model based on the fuzzy environment is used to discern the importance of different TCs,and the MADM method based on PLP(Probabilistic Linguistic Preference)-TODIM is introduced to solve the evaluator’s RIR(Relative Importance Rating).question.Finally,based on the theory of invention and innovation,the product quality is evaluated and suggestions for improvement are given.The combination of data analysis technology and decision-making model improves the deficiencies of traditional models from different aspects.From the perspective of management,this paper can not only discover users’ preferences and respond to users’ feedback quickly,but also improve the quality of products or services through the fuzzy QFD model,which can improve users’ emotional experience and brand experience,and enhance the comfort in the shopping process.At last form a win-win situation for users and manufacturers.