利用有效率之演算法探勘數量關聯規則

Using Efficient Algorithms for Mining Quantitative Association Rules

陳垂呈(Chui-Cheng Chen)

《Journal of Information Technology and Applications(資訊科技與應用期刊)》 1卷2期 (2006/09) Pp. 123-131

https://doi.org/10.6302/JITA.200609_1(2).0007

摘要

在本篇論文中，我們以包裹資料庫中之交易資料為探勘的資料來源，每一筆交易資料包含有消費者曾經購買過的項目與其數量，分別從兩方面來探勘數量關聯規則：一是修改FP-tree演算法來擷取數量關聯規則，從實驗評估中顯示，我們所提出之演算法的執行效率優於MQA-1演算法[1, 9]；二是以某一項目為探勘的目標，相較於前面的演算法，我們提出一個更有效率的演算法來擷取包含有此一項目的數量關聯規則。

關鍵字

資料探勘 ； 數量關聯規則 ； 包裹資料庫

並列摘要

In this paper, we use transaction data in bag databases as the source data of mining, and each transaction data contains a consumer ever purchased items and the quantity of those items. We mine quantitative association rules from two aspects. One is to modify the FP-tree algorithm to mine quantitative association rules. The experiments show that the performances of our algorithm are faster than the MQA-1 algorithm [1, 9]. The other is to let one transaction item as the target of mining, and to present a more efficient algorithm to mine quantitative association rules which contain the item than the preceding algorithm.

並列關鍵字

data mining ； quantitative association rules ； bag databases