| The classification of service attributes and the optimization configuration of service elements are one of the important aspects in service design or improvement.At present,researches on the classification of service attributes and the optimization configuration of service elements have attracted the attention of some scholars,and some research results have been achieved.It should be noted that,on the one hand,the data used for the classification of service attributes and the optimization configuration of service elements are mainly obtained from customers through surveys.However,surveys are expensive in terms of time and money.Besides,the quality of the data obtained from surveys depends on the complexity or length of the questionnaire and the willingness of the respondents to participate.On the other hand,with advances in information technology and Internet,customers increasingly post online reviews concerning products/services on the Internet.These online reviews contain a wealth of information.such as customers' concerns,sentiments and opinions.Relative to surveys,online reviews are not only publicly available,easily collected and low cost,but also simpler for firms to monitor and manage.Now,online reviews have been successfully used as the data source of several kinds of decision analysis,such as products ranking/recommending,customer satisfaction modelling,services improvement.brand analysis,consumer preferences analysis and market structure analysis,etc.Thus,online reviews can also serve as a promising data source for the classification of service attributes and the optimization configuration of service elements.If the the classification of service attributes and the optimization configuration of service elements can be conducted through online reviews,then it would be convenience for decision-makers or managers to obtain the classification results of service attributes and the optimization configuration results of service elements since the online reviews can be easily collected from the Internet.However,studies on conducting the classification of service attributes and the optimization configuration of service elements through online reviews have not been found.Therefore,it is necessary to study the method for classifying service attributes and optimizing the configuration of service elements based on online reviews.The purpose of this thesis is to make a deep theoretical analysis and methodological research on the classification problem of service attributes and the optimization configuration problem of service elements based on online reviews.In the view of the weaknesses of the existing research,a series of research works are conducted as follows.First,the study on the framework of the method for classifying service attributes and optimizing the configuration of service elements based on online reviews.Specifically,according to the related researches on management decision analysis through online reviews,the related concepts concerning online reviews and service attributes are first defined.Then,the framework of the methods for classifying service attributes and optimizing the configuration of service elements are given,and some descriptions of the framework are presented.Second,the study on the method for multi-class sentiment classification of online reviews concerning service attributes.Specifically,the problem of multi-class sentiment classification is studied,and a framework for multi-class sentiment classification is given.According to the framework,the experimental comparisons of feature selection and machine learning algorithms in multi-class sentiment classification are conducted.In the experiments,the performances of several commonly used machine learning algorithms and feature selection algorithms in multi-class sentiment classification are verified.On this basis,according to the obtained experimental results,a method for multi-class sentiment classification based on an improved one-vs-one(OVO)strategy and the support vector machine(SVM)algorithm is given.Finally,the effectiveness of the proposed method is verified through an experimental analysis.Third,the study on the method for identifying the Kano category of service attributes through online reviews.Specifically,the problem of identifying the Kano category of service attributes through online reviews is studied,and a framework for identifying the Kano category of service attributes through online reviews is given.According to the framework,a method for mining the useful information from online reviews is presented.Furthermore,based on the obtained useful information,a method for measuring the effects of customer sentiments toward each service attribute on customer satisfaction is proposed.On this basis,a method for identifying the Kano category of each service attribute is introduced.Finally,an empirical study is conducted to illustrate the use of the proposed method.Forth,the study on the method for conducting importance-performance analysis(IPA)of service attributes through online reviews.Specifically,the problem of conducting IPA of service attributes through online reviews is studied,and a framework for conducting IPA service attributes through online reviews is given.According to the framework,a method for mining the useful information from online reviews is presented.Furthermore,based on the obtained useful information,methods for estimating the performance and importance of each attribute are proposed.On this basis,according to the obtained attribute's performance and importance,four types of IPA plots are constructed.Finally,an empirical study concerning hotels is conducted to illustrate the feasibility and potential application of the proposed method.Fifth,the study on the method for optimizing the configuration of service elements based on the Kano and IPA classification results of service attributes.Specifically,the problem of optimizing the configuration of service elements based on the Kano and IPA classification results of service attributes is studied,and a framework for optimizing the configuration of service elements based on the Kano and IPA classification results of service attributes is given.According to the framework,a method for estimating the satisfaction degree of service elements on service attributes through online reviews is presented.Furthermore,based on the obtained results,a model for optimizing the configuration of service elements based on the Kano and IPA classification results of service attributes is given,and the solution method for the proposed model is presented.Finally,an empirical study concerning hotels is conducted to illustrate the feasibility and potential application of the proposed method.The above research work in this thesis provides a theoretical guidance framework and a methodical and technical framework to solve the problem of service attributes classification and service elements optimization configuration based on online reviews,and also lays a solid foundation to the extension and application of the related study. |
PDF Full Text Request