## AI helps you reading Science

## AI Insight

AI extracts a summary of this paper

Weibo:

# Metarule-Guided Mining of Multi-Dimensional Association Rules Using Data Cubes

KDD, pp.207-210, (1997)

EI

Abstract

In this paper, we employ a novel approach to metarule-guided, multi-dimensional association rule mining which explores a data cube structure. We propose algorithms for metarule-guided min- ing: given a metarule containing p predicates, we compare mining on an n-dimensional (n-D) cube structure (where p < n) with mining on smaller multiple...More

Code:

Data:

Introduction

- Metarule-guided mining is a interactive approach to data mining, whereby the user can probe the data under analysis by specifying hypotheses in the form of metarules, or pattern templates.
- A data mining system attempts to confirm the hypotheses by searching for patterns that match the given metarules.
- Metaruleguided mining increases the likelihood of finding rules that are of interest to the user and can make the discovery process more eficient by using the metarules to constrain the rule search space.
- Vx E person, P(x, y) A Q(x, ‘w) j buys(x, “pent&m”), (1).
- Where P and Q are predicate variables, x is a variable representing a person, and y and w are objec2 variables.
- Metarule (1 can be used to focus the data mining search towards ru 1es disclosing which two factors combined promote the sales of Pentium computers.

Highlights

- Metarule-guided mining is a interactive approach to data mining, whereby the user can probe the data under analysis by specifying hypotheses in the form of metarules, or pattern templates
- Our study explores rule mining using a data cube structure
- We propose two approaches: 1) abridged n-D cube construction, which computes the p and m-D layers of an n-D cube, while searching p-D for Lp; and 2) abridged multi-p-D cube construction, which constructs smaller, multiple p-D cubes instead of one big n-D cube for mining
- We proposed four algorithms which explore a data cube structure for metarule-guided mining of multidimensional association rules

Methods

- If the layer is larger than the available main memory, it can be partitioned into chunks (Zhao et al 1997) and only the corresponding chunks need be loaded for efficient computation.
- The data cells of the chunk are scanned, and the corresponding summary layers of the data cube are updated.
- This requires memory for the chunk being processed, and for the portions of the summary layers being updated.

Conclusion

- Previous methods for metarule-guided mining of association rules have primarily used a table-based structure, requiring costly, multiple scans of the data.
- The authors proposed four algorithms which explore a data cube structure for metarule-guided mining of multidimensional association rules.
- A hierarchy-based chunking algorithm was proposed which requires the scan of each chunk at most once and may facilitate mining of multiple-level rules

Funding

- This research is partially supported by NSERC of Canada, by NCE/IRIS, and by research grants from BC Science Council, and MPR Teltech Ltd

Reference

- Agarwal, S.; Agrawal, R.; Deshpande, P. M.; Gupta, A.; Naughton, J. F.; Ramakrishnan, R.; and Sarawagi, S. 1996. On the computation of multidimensional aggregates. In Proc. VLDB’96, 506-521.
- Agrawal, R., and Srikant, R. 1994. Fast algorithms for mining association rules. In Proc. VLDB’94,487-499.
- Chaudhuri, S., and Dayal, U. 1997. An overview of data warehousing and OLAP technlogy. SIGMOD Record 26:65-74.
- Fu, Y., and Han, J. 1995. Meta-rule-guided mining of association rules in relational databases. In Proc. Intl Workshop on Integration of Knowledge Discovery with Deductive and Object-Oriented Databases, 39-46.
- Han, J., and Fu, Y. 199Discovery of multiple-level association rules from large databases. In Proc. VLDB’95, 420431.
- Harinarayan, V.; Rajaraman, A.; and Ullman, J. D. 199Implementing data cubes efficiently. In Proc. SIGMOD’96, 205-216.
- Kamber, M.; Han, J.; and Chiang, J.Y. 199Using data cubes for metarule-guided mining of multi-dimensional association rules. CS-TR 97-10, Simon Fraser Univ, http://db.cs.sfu.ca/sections/publication/kdd/kdd.html.
- Klemettinen, M.; Mannila, H.; Ronkainen, P.; Toivonen, H.; and Verkamo, A. 1994. Finding interesting rules from large sets of discovered association rules. In Proc. CIKM’94, 401-408.
- Meo, R.; Psaila, G.; and Ceri, S. 1996. A new SQL-like operator for mining association rules. In Proc. VLDB’96, 122-133.
- Shen, W.; Ong, K.; Mitbander, B.; and Zaniolo, C. 1996. Metaqueries for data mining. In Fayyad, U.; et al. (eds), Advances in Knowledge Discovery and Data Mining. AAAI/MIT Press. 375-398.
- Sarawagi, S.; and Stonebraker, M. 1994. Efficient organization of large multidimensional arrays. In Proc. ICDE’94, 328-336.
- Srikant, R.; and Agrawal, R. 1996. Mining quantitative association rules in large relational tables. In Proc. SIG-
- Zhao, Y.; Deshpande, P. M.; and Naughton, J. F. 1997. An array-based algorithm for simultaneous multdimensional aggregates. In Proc. SIGMOD’97, 159-170.

Tags

Comments

数据免责声明

页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果，我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问，可以通过电子邮件方式联系我们：report@aminer.cn