Research on Algorithm of Function Set Mining Based on Interests

Article Preview

Abstract:

Function mining aims to discover valid functions from various data which seem not related. Nowadays, the results of most mining algorithms are not capable to characterize the things fully, because the algorithms aim at one simple function, meanwhile, another one can achieve this task, but its efficiency is very low because of the too much iteration. The paper presents one new algorithm named interest-based function set mining algorithm (IFSM), it applies the user's experience, expectations and prior knowledge to the mining process. IFSM succeeds in improving the description and enhancing the mining efficiency and quality.

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 532-533)

Pages:

1691-1694

Citation:

Online since:

June 2012

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2012 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] Candida Ferreira. Gene Expression Programming: A new adaptive algorithm for solving problems, Complex Systems, (2001).

Google Scholar

[2] Han Jia-Wei, Kambr Micheline. Data Mining—Concepts andTechniques. Beijing: Higher Education Press, 2001J.

Google Scholar

[3] ASADA T, SWONGER RF, BOUND N, et al. The quantified de-sign space: A tool for the quantitative analysis ofdesign[R]. Tech-nicalReportCMU-CS-92-213, CarnegieMellonUniversity, (1992).

Google Scholar

[4] Han J, Pei J, Yin Y. Mining frequent patt erns without candidate g ener ation [ A ] . P roc 2000 ACM-S IGMOD Int Confon Management of Data ( SIGM OD'00) [ C] . Dallas, 2000. 1-12.

DOI: 10.1145/342009.335372

Google Scholar

[5] R. Srikant, R. Agrawal. Mining Association Rules with Item Constrains. Proc. of the 3rd Int'l Conference on Knowledge Discovery in Databases and Data Mining, California, August 1997: pp.67-73.

Google Scholar