Research On Identification Of Customer Key Requirements Based On Online Reviews

Manufacturing companies are faced with increasingly fierce market competition and increasingly transparent product pricing,which makes them have to find new sources of value,and customer requirement is the source and key to product updates.With increasing international competition pressure,increasing product homogeneity,rising innovation costs,and diversifying customer requirements,more and more companies are committed to devoting limited resources to the "points" that can better capture key customer requirements Carry out product innovation.However,there are still some problems in the existing research on customer requirement mining.On the one hand,the concept of key customer requirements has not been defined,and when selecting data sources to obtain customer requirements,they are troubled by issues related to subjective interference;on the other hand,the identification of customer requirements only stays at the qualitative level and lacks a specific quantitative calculation process.In response to the above problems,this article carried out the following research:First of all,the online reviews are determined to be the data sourse for abtaining the customer requirements because of the advantages of online reviews,such as the objectivity,timeliness and easy to get.At the same time,considering the increase of the number of online reviews led the quality of the reviews declined,research on helpfulness analysis of online reviews for acquisition of customer requirements are made by bulding the elaboration likelihood model(ELM)of influence factors for helpfulness of online reviews.and based on the model,the generated variables are determined to construct the structural equation model(SEM).Eventually identified the factors that influencing the helpfulness of online reviews and their weights,and cleared which fators can really reflect and describe the customer's requirements in the online reviews,not only improve the quality of the online reviews,moreover,it further proves the rationality of identifying customer requirements based on online review data.Next,since customers' requirements are often reflected in the engineering characteristics of the product,reviews data are crawled through the web crawler,and the text data are processed by the Chinese lexical analysis system(NLIR-ICTCLAS),and then the engineering characteristics of the product are extracted by the LDA subject model algorithm.Finally,the engineering characteristics are divided into the product functional requirements to lay a foundation for of identification customer requirements.However,the Kano model is only a theoretical model with certain limitations.By referring to the concept of charismatic requirements,desired requirements and potential requirements,then the potential,but with key information,desired requirements and charismatic requirements are defined as the key customer requirements,and the model of the key customer requirements is built with the method of conjoint analysis.Combined with the analysis of sentiment and other objective factors,the customer requirements are quantitatively analyzed,and the utility value of(dis)satisfied and the key requirements index(KRI)of the product engineering characteristics are quantified,and the key degree of the customer requirements is represented from a quantitative perspective.Finally,according to the result of quantitative,further recognition level rules of key requirements are determined by the Kano model,then the final key requirements are identified,and taking the online reviews data of a smartphone as a case to identify key customer requirements,the direction of product updation are provided,and the feasibility and effectiveness of the proposed technical route and method are verified.This paper has 31 figures,21 tables,and 104 references.