臺灣博碩士論文加值系統

在現今資訊潮流中，企業的龐大資料庫可藉由統計及人工智慧的科學技術尋找出有價值的隱藏事件。利用資料做深入分析，找出其中的知識，並根據企業的問題，建立不同的模型，進而提供企業進行決策時的參考依據。資料挖掘的工作是近年來資料庫應用領域中相當熱門的議題。它雖是個神奇又時髦的技術，卻不是一門創新的學問。美國政府在第二次世界大戰前，就於人口普查以及軍事方面使用資料挖掘的分析方法。隨著資訊科技的進展，新工具的出現，以及網路通訊技術的發展，常常能超越歸納範圍的關係來執行資料挖掘，而由資料堆中挖掘寶藏，使資料挖掘成為企業智慧的一部份。在本篇論文當中，將資料挖掘技術中的關聯法則 ( Association Rule ) 運用至房地產的價格分析，進而提供有效的關聯法則，對於複雜之房價與週邊環境因素作一整合探討。購屋者將有一適當依循的投資計畫，房產業者亦可針對適當的族群做出適當的銷售企畫。

At this technological stream of time, it is able to extract the value of corporations’ large data sets by applying the knowledge of statistics and the scientific techniques from artificial intelligence. Through the use of these algorithms, the database will be analyzed and its knowledge will be generated. In addition to these, data models will be sorted by different corporation issues resulting in the reference for any strategic decision processes. More advantages are the predictions of future events and how much public is willing to contribute and feedback to new products or promotions. The probability of outcomes will be helpful as references since this information is referable to ensure companies providing quality services at the right time. In another words, companies will have clues in attempts to understand and familiarize their customers’ needs, wants and behaviors, as a result of delivering best services for customers’ satisfactions. Data mining is such a new knowledge that is commonly discussed in the field of database applications. Although it is a relatively new term, the technology is not exactly due to the analysis methods used. Before World War II, the analysis techniques were used in particular to the statistics in census or cases related to military affairs by the US government. Knowledge discovery has been one part of business intelligence in current corporations because these new techniques are inherently geared towards explicit information, rather than just simple analysis. By applying association rules from knowledge discovery technology, this dissertation will provide a discussion of price estimation in real estates. This discussion is involved in investigations into diverse housing prices resulting from the factors of surrounding environment. By referring to this association rule, buyers will acquire information about investment plans while housing agents will gain knowledge for their plans or projects in particular to their target markets.

摘要 Ⅰ
ABSTRACT Ⅱ
1.前言 1
2. DATA MINING的方法 3
2.1 DATA MINING 的簡介 3
2.2 如何利用DATA MINING作統計分析 4
3. 理論與方法 6
3.1關聯規則 6
3.2 資料分類 7
3.3 關聯法則使用方法 12
4.實例研究 17
4.1布林值的關聯規則（BOOLEAN ASSOCIATION RULE）及APRIORI演算法 18
4.2 複合維度關聯規則（MULTIPLE-DIMENSIONAL ASSOCIATION RULE） 20
4.3 關聯法則結果分析 21
5. 與線性迴歸模型的比較 23
5.1 建立迴歸模型 23
5.2 與迴歸模型的討論 25
6. 結論與未來研究方向 27
參考文獻 29
附錄-99筆樣本 -1-
附錄-45筆比對樣本 -4-

黃興進、陳啟元、周宣光、高正雄 (2005)，採用資料探勘技術建立不同醫院層級門診服務量預測模式，Journal of Taiwan Intelligent Technologies and Apply Statistics。
林傑斌、劉明德 (2006)，資料採掘與OLAP的理論與實務，台北：文魁資訊股份有限公司。
朱建平 (2006)，數據挖掘的統計方法及實踐，中國北京：中國統計出版社。
吳柏林 (2005)，模糊統計導論：方法與應用，台灣台北：五南圖書出版有限公司。
Fayyad, U. M., Piatetsky-Shapiro, G., Smyth, P. and Uthurusamy, R., editors. 1996. Advances in Knowledge Discovery and Data Mining. AAAI/MIT Press, Menlo Park, CA.
Han, J. and Fu, Y. Discovery of Multiple-Level Association Rules from Large Databases. Proc. of 1995 Int. Conf. on Very Large Data Bases（VLDB’95）, Zich, Switzerland:420-431.
Kamber. M , Han, Chiang Metarule-guided mining of multi-dimensional association rules using data cubes. BC, Canada V56A 1S6 (1997).
R. Meersman. On the complexity of mining quantitative association rules. In Data Mining and Knowledge Discovery,2,263-281,1998.
Ozden, B., S. Ramaswamy. and A. Silberschatz. 1998. Cyclic Association Rules. Proc. of 1998 Int. Conf. Data Engineering（ICDE’98）:412-421.
Piatetsky-Shapiro, G. and Frawley, W. J., editors, 1991. Knowledge Discovery in Databases, AAAI/MIT Pres, Menlo Park, CA.
T. Fukuda,Y. Morimoto, S. Morishita, and T. Tokuyama. Data mining using two-dimensional optimized association rules: Scheme, algorithms, and visualization. In Pro. ACM SIGMOD Int. Conf. Management of Data, pages13-23, Montreal, Canada, 1996.
T. Brij, G. Swinnen, K. Vanhoof, and G. Wets. Building an association rules framework to improve product assortment decisions. In Data Mining and Knowledge Discovery, pages7-23, 2004.
Y. Aumann., and Y. Lindell. A statistical theory for quantitative association rules. In Journal of Intelligent Information System, pages 255-283, 2003.