在多重支持度下挖掘數量關聯規則

Mining Quantitative Association Rules with Multiple Minimum Supports

莊嘉慧(Chia-hui Zhuang)

指導教授 ： 楊燕珠

大同大學/經營設計學院/資訊經營學系所/碩士(2005年)

摘要

過去研究所找出的數量關聯規則主要是用多量、少量等模糊值來表示， 當我們欲將項目合訂包裝促銷時，該以什麼明確的數量來搭配販售卻無法處理。以往 Chen, et al. [3] 提出 MQA-1 演算法來挖掘簡單規則，然而其所採取的是 Apriori-based 演算法易導致效率不佳，而且單一門檻值無法反映真實世界中不同商品的購買頻率不一的問題。在本文中，我們提出了一種與 FP-tree相似的結構與演算法 ( 稱為 QFP-tree )。首先，我們會延伸單一門檻值允許使用者自定多重最小支持度以反映項目的本質與購買頻率，而且我們會去區分項目被購買的次數，我們找出的關聯規則如 ”牛奶 = 3  麵包 = 2 ” ，當我們欲將牛奶和麵包合訂促銷包裝時，就可以明確的以3罐牛奶搭配2個麵包包裝成一個促銷包裏。對決策者來說，有了明確的數量資訊，決策的制定將更簡單且正確性更高。

關鍵字

關聯規則 ； 資料探勘 ； 數量關聯規則

並列摘要

The past research of mining quantitative association rules is aim to use fuzzy value like large quantity, small quantity, etc. to express quantitative attribute. It is difficult to design a bundle of items for sales promotion. In Chen’s paper [3], an Apriori-based algorithm, named MQA-1, is developed to mine association rules in bag database. However, using only one minimum support can’t reflect the nature of items.
In this paper, we propose a FP-tree-like structure to store all information about itembag and an efficient algorithm to mine quantitative association rules with multiple minimum supports. It form looks like “milk = 2  bread = 3”, then we can combine three units of milk with two units of bread to form bundling. For decision makers, it is easy and precise to make decisions with quantity information.

並列關鍵字

association rule ； data mining ； quantitative association rule

參考文獻

[3]. Chen, Y.L., Hsu, P.Y. and Ling, C.C. “Algorithms for mining association rules in bag databases,” Information Science Volume: 166, Issue: 1-4, 2004, pp. 31-47.

[6]. Hu, Y. H. and Chen, Y. L. “Mining association rules with multiple minimum supports: a new mining algorithm and a support tuning mechanism,” forthcoming in Decision Support Systems.

[7]. Kuok, C.M., Fu, A. and Wong, M.H. “Mining Fuzzy Association Rules in Databases,” ACM SIGMOD, 1998, pp. 41-46.

[1]. Agrawal, R., Imielinski, T. and Swami, A. “Mining Association Rules between Sets of Items in Large Databases,” Proceedings of the 1993 ACM SIGMOD international conference on Management of data, 1993, pp. 207-216.

[2]. Agrawal, R. and Srikant, R. “Fast algorithms for Mining Association Rules,” Proceedings of the 20th International Conference on Very Large Data Bases, 1994, pp. 487-499.