企業與組織之間為了提升營運能力，因此對決策支援系統的需求量大增，目前的系統皆針對特定議題進行，並無一套較廣泛性、一般化應用的系統，然而由於大數據成為目前主流的趨勢，資料搜集變得非常容易，也因此可分析議題也隨之暴增，幅度甚至超過所蒐集資料量的增加，因為可能關注的分析議題，是與資料間的各種排列組合成正比(或更快速成長)；而資料的排列組合相對於資料量的大小，是成指數程度的成長。決策支援系統需要科學化論證來說明此決策預期達到的成效，因此本研究以學生為主題切入，不分議題的進行廣泛探索性分析，找尋具有因果關係的資料組合，並將因果關係強度加以量化，以證明此決策的成效，此外更納入多維度與多階層資料架構，挖掘出多樣化、多觀點的因果規則，並從因果規則中推論間接因果規則，釐清整件事情的前因後果。本研究的目標為找出因果規則提供研究人員分析議題發想，亦可提供多元化的決策支援。;In order to improve the operating abilities, enterprises and organizations have greatly increased their demand for decision support systems. The current systems are all focused on specific issues, and there is not exist a more extensive and general system. However, big data has become the current trends. Data collection has become very easy. Therefore, the number of analyzable topics has also increased sharply. Even exceeding the increase of the amount of collected data. Because the analysis topics may be proportional to the various permutations and combinations of the data (or more rapid growth). And the permutation and combination of data grow exponentially with dataset size. The decision support system requires scientific argumentation to illustrate the expected results Therefore, this research takes students as the theme and conducts extensive exploratory analysis. Find out the combinations of data which has causal relationship and quantifies the strength of the relationship to prove the effectiveness of decision-support. Our research also constructs a multi-dimensional and multi-level data structure to mining diversified and multi-view causal rules. Also infer indirect causal rules from the causal rules to clarify the cause and effect. Our goal is to find out causal rule for decision-support and also provide new research topics for researchers.
