| In the fierce market competition,accurate evaluation of products or services and targeted improvement strategies have become the key for enterprises to gain advantages in the competition.The researches on product evaluation and improvement methods have attracted the attention of many scholars and the related researches have been applied in many fields.However,most of the traditional product evaluation and improvement methods are based on the form of survey or interview to collect data,with high cost,easy to outdated and other disadvantages.With the continuous development of social media and the arrival of the era of big data,online reviews given by consumers have become ubiquitous.Many scholars have conducted rich researches on online reviews.Existing studies have shown that online reviews contain a wealth of consumer information,such as concerns,preferences,needs and sentiment information.Online reviews have become an important data source for the practical decisionmaking problems.However,the research on product evaluation based on online reviews is just beginning.A small number of product evaluation and improvement studies on online reviews have deficiencies such as simple online reviews processing and insufficient comprehensive consideration.Therefore,the method of product evaluation and improvement based on online reviews still needs futher exploration.On the basis of summarizing and analyzing the previous methods,this thesis conducts indepth research on product evaluation and improvement based on online reviews,and mainly carries out the following three aspects of research:(1)A product evaluation and improvement method considering the enterprise's own online reviews is proposed.This method establishes an improved Kano model by analyzing the useful consumer information contained in the enterprises' own online reviews,so as to provide efective product evaluation for the enterprise and provide suggestions for product improvement under the condition of limited resources.Firstly,the online reviews on the online website are obtained,and the product attributes are extracted and sentiment analysis is carried out on the online reviews.Secondly,the traditional Kano model is extended,and the sentiment information reflected by online reviews is used to identify attribute types.Then,according to the influence of different attribute types on consumer satisfaction,the satisfaction functions of different attributes are calculated.Finally,three aspects of attribute types,attribute performances and satisfaction functions were considered comprehensively,and suggestions for product improvement were put forward.(2)A product evaluation and improvement method considering competitor's online reviews is proposed.This method improves the Importance-Performance Analysis(IPA)model by considering the influence of competitors and attribute types,and provides management suggestions for enterprises to improve their market competitiveness through product evaluation and improvement.Firstly,based on the online reviews,the comparison information between the enterprise and its potential competitors is extracted.Secondly,identify the main competitors based on comparative information.Then,the improved Kano model is used to distinguish attribute types,and the importance of attributes is redefined by considering the characteristics of different attribute types,so as to improve the IPA model.Finally,according to the results of the improved IPA model,suggestions for product improvement are analyzed.(3)A product evaluation and improvement method based on online reviews of market environment is proposed.By considering the influence of the overall market environment,this method establishes a product evaluation model and provides suggestions for enterprises to make product improvement more in line with the market demand.First of all,the online reviews of multiple products in the market is obtained,and the product attributes and corresponding sentiment information of consumers are obtained through attribute extraction and sentiment analysis.Secondly,the improved Kano model is used to judge the attribute types and set different market reference points according to the attribute types.Then,based on the market reference points of each attribute,the gain and loss situation of the enterprise's product attribute performances relative to the market reference points are calculated,and the ranking of the enterprise's product in the market is calculated according to the gain and loss situation.Finally,according to the gains and losses of each attribute of the product and the ranking results,the product improvement suggestions are put forward,and the effectiveness of the method are illustrated by a real case of hotel service improvement and the simulation analysis of correlation coefficients.The product evaluation and improvement methods based on online reviews proposed in... |
PDF Full Text Request