在知識經濟的時代,除了分享、應用社群中的知識外,知識的發現也成為了一個很重要的議題,而資料探勘在知識發現的過程裡扮演了一個很重要的角色,在此本研究嘗試一個新的資料探勘架構:先叢集分析,再進行關聯式法則的探勘;透過此一架構,期望能以最小的誤差下,達成提升資料探勘效能,使得資料探勘更容易應用於真實世界上。 本研究首先將醫療資料庫中的疾病代碼透過國際疾病分類碼(ICD-9-CM)的分類縮減資料的維度,再透過自組織映射圖(SOM)進行集群分析,最後再利用以螞蟻理論為基礎的關聯法則探勘方式,對各集群進行關聯法則的探勘,發現透過此一資料探勘架構,不但可以提升探勘的效率,透過集群的分析,可以輕易的探索某特定一部分的關聯法則,以利在面對龐大資料下,可以更輕易的掌握到有用的知識。
In addition to sharing and applying the knowledge in the community, knowledge discovery has become an important issue in the knowledge economic era. Data mining plays an important role of knowledge discovery. Therefore, this study intends to propose a new framework of data mining that does clustering analysis first, and then followed by association rule mining. The study reduced the data dimensions by the classifications of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) first, and then clustered the data set with Self-organizing Map(SOM) network. Finally, we mined the association rule in all clusters by ACS-based association rule mining system. The result showed that the new mining framework can provide not only the better effect, but also the easier way to find the useful rules that maybe hidden in the very large data. In other words, it is easier to extract the useful knowledge by the proposed framework.