評論關注點、評論傾向與內容代表性之綜合評估模式

An Model for Comment Value Evaluation

藍玉潔(Yu-Jie Lan)

指導教授 ： 侯建良

國立清華大學/工學院/工業工程與工程管理學系/碩士(2018年)

摘要

當資訊需求者欲掌握其感興趣產品/服務之評價時，其往往透過各網路論壇瀏覽與該產品/服務相關之評論。而資訊需求者往往依其個人偏好挑選評論進行閱讀，忽略其餘評論內容。此外，資訊需求者往往將各評論內容中提及之產品/服務之各關注項目的評價水準以同等代表性程度進行評價彙整，而忽略各評論內容之代表性程度不同、其對產品/服務之各關注項目評價彙整的影響程度實不相同，導致資訊需求者無法適切地彙整各則具不同代表性程度的評論所提及之產品/服務之各關注項目的評價水準。為解決此問題，本研究首先解析自各網路論壇蒐集之產品/服務相關評論，並釐清各評論內容所包含之特徵屬性（包括評論發佈日期、留言者等級、留言者涉入程度等）。接著，依前述解析結果，本研究乃發展一套「評論關注點、評論傾向與內容代表性之綜合評估模式」方法論，此方法可擷取各評論內容中之特徵屬性，並依五項代表性相關特徵屬性分析一評論之評價代表性。之後，此方法乃將所有評論內容探討之產品/服務關注項目評價水準依不同評論之代表性程度進行評價彙整，以產生一產品/服務關注項目之評價結論，並以視覺化方式將評價趨勢及評價結論呈現予資訊需求者，以利資訊需求者準確作出最終消費決策。

關鍵字

評論評價評估 ； 評價彙整 ； 分群方法

並列摘要

Once a customer wants to capture critical information from the comments for a product/service, he/she has to search for the comments related to the product/service via the Internet. As the customer browses the comments, he/she usually has to spend much time to acquire the emotional tendency of the product/service. Furthermore, the customer often regards that the value of each comment is of equal importance. That is, the representativeness of distinct comments is usually ignored. In order to solve this problem, this research develops a model for comment value evaluation. The proposed model can be used to visually reveal the values of distinct comments and the integrated evaluation of the product/service. By utilizing this model, the customer can easily acquire representativeness of each comment and the integrated evaluation of product/service characteristics in order to efficiently make purchase decisions.

並列關鍵字

comment evaluation ； value integration ； cluster method

參考文獻

1. Abrahams, A. S., Jiao, J., Wang, G. A. and Fan, W., 2012, “Vehicle Defect Discovery from Social Media,” Decision Support Systems, Vol. 54, pp. 87-97.

2. Aramaki, E., Maskawa, S. and Morita, M., 2011, “Twitter Catches the Flu: Detecting Influenza Epidemics Using Twitter,” Proceedings of the Conference on Empirical Methods in Natural Language Processing, pp. 1568-1576.

3. Azmandian, M., Singh, K., Gelsey, B., Chang, Y. H. and Maheswaran, R., 2012, “Following Human Mobility Using Tweets,” International Workshop on Agents and Data Mining Interaction, pp. 139-149.

4. Barnett, G. A., Xu, W. W., Chu, J., Jiang, K., Huh, C., Park, J. Y. and Park, H. W., 2016, “Measuring International Relations in Social Media Conversations,” Government Information Quarterly, pp. 1-8.

5. Bello, G., Menendez, H., Okazaki, S. and Camacho, D., 2013, “Extracting Collective Trends from Twitter Using Social-Based Data Mining,” International Conference on Computational Collective Intelligence, pp. 622-630.