摘要: | 校務研究目前在世界各國受到許多學校的重視,台灣近年來因為大專院校不 斷增設,衍生出許多問題,為了讓決策者可以從不同面向分析並做出最合適的決 策,政府機關與各高等教育機構亦開始極力推行校務研究,以更科學化論述事證 的方式面對環境與教育政策的改變。 校務研究涵蓋的範圍極為廣泛,包含學務、教務、人事以及資源等,目前校 務研究單位雖整合了學校各行政單位的資料,但仍有許多有價值的資料尚未被分 析和使用。本研究以學生為主要分析面向,每位學生背景不同,在學時參與的活 動、修習及獎懲等表現也不同,學生在校的歷程往往會對畢業後的發展趨勢有所 影響。藉由學生入學前的入學方式與高中類型,以及在學的社團參與、獎懲、工 讀等加上畢業的工作類型與薪資等分析學生發展的趨勢,對於學校評估學生學習 成效或是輔助校務決策有許多幫助。 為了由不同面向發掘資料的因果規則,本研究以多維度資料倉儲作為分析的 資料來源,將不同維度的屬性依照時間發生的順序形成序列,再透過循序樣式探 勘,探索學生在校狀況及畢業流向間有效的因果規則,以支援校務決策之重要分 析。;Institutional Research has been a significant feature of higher education in other countries for many years. In Taiwan, as the education policy and society environment change rapidly in responding to pressures of globalization and competition, decision makers need scientific methods to catch up with the times. Therefore, higher education institutions are starting to implement institutional research these years. The wide coverage of topic of institutional research includes academic affair, student affair, personnel affair, general affair, etc. In IR data warehouse, integrated data can be turned into meaningful information for decision making. This paper focuses on finding causal rules from student portfolios including type of admissions, clubs, rewards and punishments, part-time jobs and questionnaire survey of graduation to analyze career development. In order to support decision making, in this paper, data from multi-dimensional data warehouse are seen to be employed as input data and an improved sequential pattern mining method is proposed for discovering useful causal relations. |