Presentation 2005-07-14
Data Analysis Method for Extracting Multiple Ratio Rules
Masafumi HAMAMOTO, Hiroyuki KITAGAWA, Christos FALOUTSOS,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Ratio Rules are correlation among attribute values, and are applicable to data analysis, value estimation, and so on. Many Ratio Rules are generally included in data; we call them Multiple Ratio Rules. To extract each Ratio Rule in Multiple Ratio Rules, we consider the Apriori property for Ratio Rules. This enables us to extract any Ratio Rules by using two dimentional Ratio Rules. To extract two dimentional Ratio Rules, we divide tuples into small areas called buckets and extract Ratio Rules from the histogram, the number of tuples in buckets. We examine our proposed method using synthetic data and validate its usefulness.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Ratio Rule / Apriori property / Data Mining
Paper # DE2005-71
Date of Issue

Conference Information
Committee DE
Conference Date 2005/7/7(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 Data Engineering (DE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Data Analysis Method for Extracting Multiple Ratio Rules
Sub Title (in English)
Keyword(1) Ratio Rule
Keyword(2) Apriori property
Keyword(3) Data Mining
1st Author's Name Masafumi HAMAMOTO
1st Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba()
2nd Author's Name Hiroyuki KITAGAWA
2nd Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba:Center for Computational Sciences, University of Tsukuba
3rd Author's Name Christos FALOUTSOS
3rd Author's Affiliation Carnegie Mellon University
Date 2005-07-14
Paper # DE2005-71
Volume (vol) vol.105
Number (no) 172
Page pp.pp.-
#Pages 6
Date of Issue