命理在華人社會中一直有很大的影響力，近年來隨著使用網際網路的人口快速增長，在台灣命理相關網站日漸風行。本研究使用資料探勘工具，針對台灣民眾使用網路算命行為，分析網路算命使用者的個人特質、使用行為以及使用者動機。本研究以約略集合理論（Rough Set Theory）、決策樹（Decision Tree）與貝氏網路（Bayesian Net-work）的資料探勘方法來進行分析，希望能區隔民眾使用網路算命之特徵、偏好，以瞭解台灣民眾使用網路算命的趨勢。研究發現：網路算命使用者與其個人宗教信仰以及網路宗教的使用密切關係，有朝向上班族及高學歷兩種趨勢，本研究結果將有助於增進對網路算命研究課題之瞭解，同時也可以提供網路業者做為經營參考。

Fortune telling has always been influential in Chinese society. In recent years, internet users have increased considerably in number. This has led to attention to internet-related behavior and characteristics. This study aims to understand the motivation and behavior pattern for Taiwanese users of on-line fortune telling. This paper proposes a method that links the rough set theory, Bayesian network and decision tree modeling for causal analysis. According to our research, usage of on-line fortune telling is related to users' religion and online religion experience. On the other hand, there are a growing number of well-educated people and full-time workers who use online fortune telling. Data mining results for targeting marketing to on-line users can offer Internet enterprises information for marketing decision making. Based on the findings of this study, conclusions and implications for management are discussed.

網路算命；資料探勘；約略集合理論；決策樹；貝氏網路

online fortune telling； Data Mining； Rough Set Theory； Bayesian network； Decision Tree