評論語意自動化分析與顧客滿意度應用之研究-以中國旅遊網站為例

A Research of Automated Comments Semantic Analysis and Its Application in Customer Satisfaction – A Case Study on a Chinese Tourism Website

許榮望(Rong-Wang Hsu)

指導教授 ： 李麗華

朝陽科技大學/資訊學院/資訊科技研究所博士班/博士(2013年)

摘要

在E世代環境下，顧客至雲端發表對產品或服務的評論或滿意度的回應，已是網路常見的一種表達形式。但是，在這種資訊大量充斥的網路環境下，企業分析顧客留言資訊或評論以做為產品或服務滿意度的參考依據，大都仍使用人工方式來處理大量的資料。顧客對服務品質的滿意度研究大多仍著重在評價(rating)的量化統計分析，或透過問卷調查方式，鮮少利用顧客評論(comment)的資料做語意分析。但是，只有量化資料，是無法使企業具體了解顧客在意的項目內容或服務需要改進的部分，而使用人工處理大量資料或問卷調查方式不但會耗費很多的人力與時間，問卷資料來源的可靠性與正確性也頗受爭議。

由於現今雲端及語意網的盛行，如何善用顧客評論，進行語意自動化分析，勢必成為企業非常重要的利器。因此，本研究第一部分先利用顧客的旅館評論資料，運用本體論架構建置評論語意字彙，藉由字彙比對的結果，做為語意轉換評價值的方式。最後由語意分析轉換後的評價值與顧客給予的評價值分別對旅館服務品質的滿意度進行分析。經由檢驗結果證實本研究所提之評論語意自動化分析方法確實具有極高的可應用性。

本研究第二部分則對於評論語意轉換後的評價值深入分析旅館服務品質構面對顧客滿意度之影響。最後經由本研究之分析證實：(1)服務品質與顧客滿意度均有顯著的影響(反應性除外)，而且以可靠性、有形性、關懷性等三構面特別顯著。(2)旅館星級與價格對服務品質與顧客服務品質滿意度均有顯著性的影響(反應性除外)。旅館價格具有負向調節效果，旅館星級則有正向調節效果。反應性對於星級和價格則沒有顯著性的影響。

關鍵字

本體論 ； 語意分析 ； 評論 ； 評價 ； 滿意度 ； 服務品質

並列摘要

In the era of E-generation today, comments on products or services or satisfaction feedback by customers have become a common method of expression on the Internet. However, given the massive amount of information available in the Inter-web environment, most businesses still rely on manual processing of massive information in the analysis of customer feedback or comments as the basis of product or service satisfaction performance. In fact, most studies on customers’ satisfaction of service quality still focus on quantitative statistical analysis of ratings or questionnaires and few researches have presented semantics analysis of customer comments. The problem with quantified data lies in the fact that businesses would not be able to attain concrete understanding of the contents that customers really care about or aspects of their services that require improvement. Manual processing of massive information or questionnaires is not only time and labor consuming but also offers limited reliability and accuracy when it comes to the sources of questionnaires.

Given the popularity of cloud computing and semantic web today, adequate use of customer comments for automated semantic analysis would no doubt become an important tool for service providers. Therefore, the first part of this study aims to present a model of comment semantic words built on ontological theories to perform semantic conversion of customer comments on hotel services to ratings through the process of word matching. Ultimately, ratings given by the customers and ratings derived from the conversion of semantic analysis would be used for the analysis of customers’ satisfaction on hotel service quality. The method proposed in this study has proven that semantic analysis is capable of delivering results of high applicability.

The second part of this study about the comment semantic evaluation values converted in-depth analysis the hostel service quality has the effect of customer satisfaction. This study’s analysis has proved: (1) There is significant influence between service quality and customer satisfaction (except for responsiveness), especially in terms of reliability, tangibles and empathy. (2) The hotel start and price have a significant impact on service quality and customer satisfaction (except for responsiveness). The hostels price has negative moderating effect, hotel star positive moderating effect. Responsiveness for the star and the price is not a significant impact.

並列關鍵字

Ontology ； Semantic Analysis ； Comment ； Rating ； Satisfaction ； Service Quality

參考文獻

[98] Hempel, D. J. (1997), “Consumer Satisfaction with the Home Buying Process: Conceptualization and Measurement,” Marketing Science Institute, Vol. 7, No. 4, pp. 377-399.

[1] Bickart, B. & Schindler, R. M. (2001), “Internet Forums as Influential Source of Consumer Information,” Journal of Interactive Marketing, Vol. 15, No. 3, pp. 31-40.

[2] Parasuraman, A., Zeithaml, V.A., & Malhotra, A. (2005), “E-S-QUAL: A Multiple-Item Scale for Assessing Electronic Service Quality,” Journal of Service Research, Vol. 7, No. 3, pp. 213-233.

[4] Tanimoto, J. & Fujii, H. (2003), “A Study on Diffusion Characteristics of Information on a Human Network Analyzed by a Multi-Agent Simulator,” The Social Science Journal, Vol. 40, No. 3, pp. 479-485.

[7] Schuman, H. & Presser S. (1981), “Questions and Answers in Attitude Surveys: Experiments on Question Form, Wording, and Context,” New York: Academic Press, pp. 23-77.