Presentation | 2003/3/6 An Efficient Mining Method for Episode Rules using Approximate Informative Basis Yusuke FUJITA, Makoto HARAGUCHI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Discovery of association rules from time-series datasets is an important data mining task. Generally, the number of potential rules grows rapidly as the size of database increases. It is therefor hard for a user to analyze the rules and realize useful ones among them. To avoid such a difficulty, we make some rules invisible to users, provided they are redundant and approximately reconstructed from another non-redundant ones. In another words, only non-redundant rules are presented to users and will be checked for their interestingness. For this purpose, we first define a notion of approximate informative basis consisting of only non-redundant rules, and then present an efficient method to construct it. The degree of approximate reconstruction is associated with the basis as a real parameter adjustable by users. Our experimental results show that the number of non-redundant rules in the approximate informative basis is much reduced. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | association rule mining / time-series dataset / vent sequence / episode rule / approximate redundancy |
Paper # | AI2002-65 |
Date of Issue |
Conference Information | |
Committee | AI |
---|---|
Conference Date | 2003/3/6(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Artificial Intelligence and Knowledge-Based Processing (AI) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | An Efficient Mining Method for Episode Rules using Approximate Informative Basis |
Sub Title (in English) | |
Keyword(1) | association rule mining |
Keyword(2) | time-series dataset |
Keyword(3) | vent sequence |
Keyword(4) | episode rule |
Keyword(5) | approximate redundancy |
1st Author's Name | Yusuke FUJITA |
1st Author's Affiliation | Division of Electronics and Information Engineering Hokkaido University() |
2nd Author's Name | Makoto HARAGUCHI |
2nd Author's Affiliation | Division of Electronics and Information Engineering Hokkaido University |
Date | 2003/3/6 |
Paper # | AI2002-65 |
Volume (vol) | vol.102 |
Number (no) | 709 |
Page | pp.pp.- |
#Pages | 5 |
Date of Issue |