Publications

Detailed Information

Evaluation of Online Services Using the MCDM Approach: User Evaluation and Expert Evaluation : 다기준 의사결정 방법론을 활용한 온라인 서비스의 평가: 사용자 평가와 전문가 평가

DC Field Value Language
dc.contributor.advisor박용태-
dc.contributor.author강대국-
dc.date.accessioned2017-07-13T06:06:16Z-
dc.date.available2017-07-13T06:06:16Z-
dc.date.issued2015-08-
dc.identifier.other000000066729-
dc.identifier.urihttps://hdl.handle.net/10371/118272-
dc.description학위논문 (박사)-- 서울대학교 대학원 : 산업·조선공학부, 2015. 8. 박용태.-
dc.description.abstractThe recent development of the Internet technology has drawn attention toward online services. In response, many kinds of online services have been developed. These online services have some distinct characteristics compared with those of traditional offline services. First, online services can provide more service functions than offline services, because there is no restriction of time and space in terms of the service provision. Second, customers often have deficient information about online service providers, because online service provision generally takes place between parties who have never transacted with each other before. Third, online services have difficulty providing the spontaneous responses that can be offered in offline services through face-to-face contact. Given these circumstances, the evaluation of online services through continuous monitoring to reliably provide various service functions has been emphasized. In response, much attention has been paid to the evaluation of online services. Many studies have utilized multiple criteria decision-making (MCDM) approaches for evaluating online services by considering various evaluation criteria simultaneously. Although previous studies contributed to improving our initial understanding of the evaluation of online services, they were limited, as they did not consider the characteristics of evaluators and different types of evaluation data.
Accordingly, the focus of this study is to suggest an evaluation framework of online services that considers characteristics of evaluators and types of data in the process of evaluation. The study is composed of two modules and three themes. First, module 1 focuses on the issue of user evaluation of online services using MCDM. Specifically, user evaluation approaches can be divided into two types in terms of the data that can be collected from evaluators: evaluations based on customer reviews and evaluations based on surveys. Therefore, different frameworks for user evaluation in terms of data are provided as different themes. Second, module 2 deals with the issue of expert evaluation of online services using MCDM. This module includes the third theme. The objectives of the three themes are specified as follows.
Theme #1 aims to suggest a framework for user evaluation of online services using customer reviews. Customer reviews have been recognized as valuable information sources for monitoring and enhancing customer satisfaction levels, and they provide the real voices of users. Each customer review has concise core content focused on important criteria in comparison with survey data, which contains information about evaluation scores for alternatives in terms of various criteria. It can be useful in situations in which we cannot collect survey data. To analyze a large number of customer reviews effectively and evaluate online services, this study combines VIKOR (in Serbian: ViseKriterijumsa Optimizacija I Kompromisno Resenje) and sentiment analysis. The suggested framework of this study mainly consists of two stages: (1) data collection and preprocessing and (2) customer satisfaction measurement. In the first stage, data collection and preprocessing, text mining is utilized to compile customer-review-based dictionaries of attributes and sentiment words. Then, using sentiment analysis, sentiment scores for attributes are calculated for each mobile service. In the second stage, levels of customer satisfaction are measured using VIKOR. For the purpose of illustration, an empirical case study was conducted on customer reviews of mobile application services. Theme #1 contributes to propose a systematic way of utilizing customer reviews in the evaluation process
Theme #2 presents a tailored framework for user evaluation of online services using survey data. In this study, fuzzy hierarchical TOPSIS based on E-SERVQUAL (E-S-QUAL) is suggested for the effective evaluation of online services using information gathered from surveys. To put it more concretely, with E-S-QUAL, which is the extended version of SERVQUAL for the measurement of electronic service quality, information about the gap between the expected and perceived service level collected from users can be extracted effectively. In addition, with fuzzy hierarchical TOPSIS, the hierarchical structure of the evaluation criteria between the main dimensions and their sub-criteria can be considered. With this approach, based on the preservation of the core concept of E-S-QUAL, the hierarchy between the main dimensions and their sub-criteria can be captured.
Theme #3 focuses on the suggestion of an integrated model consisting of fuzzy DEMATEL and fuzzy VIKOR as a stepwise method of expert evaluation of online services. This theme mainly focuses on the effective approach for making the best use of knowledge of expert opinions collected from surveys. For expert evaluations, more detailed information (e.g., interrelationships among evaluation criteria) is collected. The evaluation process proceeds as follows. First, several online services providing similar functions are selected as alternatives, and service-based and product-based criteria are defined based on a literature review and expert opinions. Second, interrelationships and importance weights of criteria are calculated using fuzzy DEMATEL. Third, the online service is evaluated using fuzzy VIKOR. In this theme, the inevitable and often conflicting relationships among service-based criteria and product-based criteria can be effectively captured to reflect the evaluation of online services.
-
dc.description.tableofcontentsContents

Chapter 1. Introduction 1
1.1. Background and motivation 1
1.2. Purpose 3
1.3. Scope and framework 5
1.4. Thesis outline 6
Chapter 2. Background 8
2.1. Theoretical background 8
2.1.1. Evaluation of online services 8
2.1.2. MCDM approach 10
2.2. Methodological background 11
2.2.1. Sentiment analysis 12
2.2.2. VIKOR 13
2.2.3. E-S-QUAL method 15
2.2.4. Fuzzy set theory 17
2.2.5. Fuzzy hierarchical TOPSIS 21
2.2.6. Fuzzy DEMATEL method 22
Chapter 3. User evaluation of online services using MCDM 23
3.1. User evaluation based on customer reviews 23
3.1.1. Introduction 23
3.1.2. Proposed approach 26
3.1.3. Empirical case study 33
3.1.4. Conclusions 41
3.2. User evaluation based on surveys 43
3.2.1. Introduction 43
3.2.2. Proposed approach 47
3.2.3. Empirical case study 57
3.2.4. Conclusions 73
Chapter 4. Expert evaluation of online services using MCDM 75
4.1. Expert evaluation based on surveys 75
4.1.1. Introduction 75
4.1.2. Proposed approach 77
4.1.3. Empirical case study 85
4.1.4. Conclusions 98
Chapter 5. Conclusions and discussions 99
5.1. Summary and contributions 99
5.2. Limitations and future research 101
Bibilography 103
초록 116


List of Tables

Table 2 1 Perspectives on evaluation of online services 9
Table 2-2 Methodologies utilized in this study 12
Table 2 3 Descriptions of four dimensions of E-S-QUAL 16
Table 2 4 TFNs of linguistic variables for rating of alternatives 19
Table 2 5 TFNs of linguistic variables for importance weight of criteria 20

Table 3 1 Number of customer reviews in each category of mobile application service 34
Table 3 2 Criteria for measurement of customer satisfaction with mobile application service in category of social networking service 35
Table 3 3 Calculated scores of attributes in each mobile application service 36
Table 3 4 Normalized scores of attributes and weights of attributes 37
Table 3 5 Positive ideal solutions and negative ideal solutions 37
Table 3 6 Scores of 38
Table 3 7 values and ranking 39
Table 3 8 Evaluation criteria of four dimensions and sub-criteria of E-S-QUAL 58
Table 3 9 TFN of linguistic variables for rating of alternatives 60
Table 3 10 Aggregated expected fuzzy values and the perceived fuzzy values 61
Table 3 11 Defuzzified matrix based on calculated BNP values 62
Table 3 12 Aggregated normalized decision matrix 64
Table 3 13 Importance weights of main dimensions of E-S-QUAL 65
Table 3 14 Importance weights of sub-criteria of E-S-QUAL 66
Table 3 15 Final importance weights of sub-criteria of E-S-QUAL 67
Table 3 16 Weighted normalized decision matrix 68
Table 3 17 Final weighted normalized decision matrix by addition principle 69
Table 3 18 PIS and NIS for each alternative 69
Table 3 19 Distance from PIS and NIS for each alternative 70
Table 3 20 Closeness coefficients ( ) of six alternatives and ranking order 71
Table 3 21 Results of the sensitivity analysis 72

Table 4 1 Criteria for mobile service evaluation within navigation category 86
Table 4 2 Initial direct-influence fuzzy matrix as assessed by evaluators 87
Table 4 3 Aggregated initial direct-influence defuzzified matrix 88
Table 4 4 Total influence matrix 89
Table 4 5 Linguistic variables for rating of alternatives 92
Table 4 6 Aggregated defuzzified ratings of alternatives 93
Table 4 7 Positive ideal solutions and negative ideal solutions 93
Table 4 8 Scores of 94
Table 4 9 values and ranking 95


List of Figures

Figure 1-1 Purposes of two modules 3
Figure 1-2 Type of evaluation in this study 5
Figure 1-3 Scope of study 6
Figure 1-4 Overall structure of study 7

Figure 2 1 Membership functions of linguistic variables for rating 19
Figure 2 2 Membership functions of linguistic variables representing importance weight of criteria 19

Figure 3 1 Overall research framework 27
Figure 3 2 Example of construction of keyword vector 30
Figure 3 3 Examples of normalized polarity scores for attributes 32
Figure 3 4 Dictionary of sentiment words 35
Figure 3 5 Sensitivity analysis 40
Figure 3 6 Research framework for evaluation of B2C e-commerce websites 47
Figure 3 7 Results of the sensitivity analysis 73

Figure 4 1 Overall framework 78
Figure 4 2 Influence relation map (IRM) 90
Figure 4 3 Sensitivity analysis 97
-
dc.formatapplication/pdf-
dc.format.extent33882553 bytes-
dc.format.mediumapplication/pdf-
dc.language.isoen-
dc.publisher서울대학교 대학원-
dc.subjectOnline service evaluation-
dc.subjectuser evaluation-
dc.subjectexpert evaluation-
dc.subjectmultiple criteria decision-making (MCDM)-
dc.subjectsentiment analysis-
dc.subject.ddc623-
dc.titleEvaluation of Online Services Using the MCDM Approach: User Evaluation and Expert Evaluation-
dc.title.alternative다기준 의사결정 방법론을 활용한 온라인 서비스의 평가: 사용자 평가와 전문가 평가-
dc.typeThesis-
dc.contributor.AlternativeAuthorDaekook Kang-
dc.description.degreeDoctor-
dc.citation.pagesviii, 118-
dc.contributor.affiliation공과대학 산업·조선공학부-
dc.date.awarded2015-08-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share