陳怡廷、欒錦榮

Yi-Ting Chen,Chin-Jung Luan

中文斷詞 ； 文字探勘 ； 自然語言處理 ； 詞語共現網絡 ； 網路口碑

Chinese knowledge information processing ； text mining ； natural language processing ； term co-concurrence network ； on-line word-of-mouth

近年，配合資訊處理技術進步，研究者開始可以利用電腦軟體系統將文本數據資料庫化，進行系統化的分析。本研究試圖透過自然語言處理，建立一可將網路口碑內容資料庫化的分析處理程序。藉由中研院中文斷詞系統將網路口碑內容系統化地處理分析，並透過質性分析軟體長時間地記錄、存儲，用以建置一可永續發展且能提供行銷或消費者行為研究應用的資料庫平台。

This study establishes a research method for archiving the issues of online word-of-mouth (WOM) using the Chinese knowledge information processing (CKIP) and natural language processing (NLP) methods. Online WOM is an important self-expression medium that can ascertain customers' attitudes and assessments for specific product(s) or service(s), enabling an understanding of their direct responses to products. Using this research method to study a leading mobile phone manufacturer, we plotted the term co-concurrence networks to obtain meaningful marketing implications and to demonstrate the usefulness of creating an efficient method for building a database platform that consists of systematically recorded, archived, and processed data for marketing researchers and practitioners.