Presentation 2001/3/12
A Data Preprocessing Mechanism Based on Processing Attribute values and Sellecting Attributes
Mao KOMORI, Hidenao ABE, Yoshiaki TACHIBANA, Takahira YAMAGUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper discuss a methodology for data preprocessing in KDD, forcusing on the construction of a set of good attributes. We start with a set of seed attributes that come uo frequency in data mining result in advance. They can be extended with other attributes while the incremental extention process provids mining result better than just before. We also do the following processing of attribute values : two convert numeric values to symbolic values or a symbolic values to group values for the purpose of reducing the serch space of irrelevant attributes, in order to evaluate our methodology we take a case study from baseball datasets. Five important attributes have been selected as seed as attributes and they can be extended with three other atributes. The eight attributes have shown best performance. All over the combination of attributes so our methodlogy van work over this small size case study. We will generalize and scall up our methodlogy to other large size of data sets.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) knowredge discovery / data pre-processing / feature selection / discrete / sports science
Paper # AI2000-77,KBSE85
Date of Issue

Conference Information
Committee AI
Conference Date 2001/3/12(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) A Data Preprocessing Mechanism Based on Processing Attribute values and Sellecting Attributes
Sub Title (in English)
Keyword(1) knowredge discovery
Keyword(2) data pre-processing
Keyword(3) feature selection
Keyword(4) discrete
Keyword(5) sports science
1st Author's Name Mao KOMORI
1st Author's Affiliation Dept. Computer Science, Shizuoka Univ.()
2nd Author's Name Hidenao ABE
2nd Author's Affiliation Dept. Computer Science, Shizuoka Univ.
3rd Author's Name Yoshiaki TACHIBANA
3rd Author's Affiliation Dept. Computer Science, Shizuoka Univ.
4th Author's Name Takahira YAMAGUCHI
4th Author's Affiliation Dept. Computer Science, Shizuoka Univ.
Date 2001/3/12
Paper # AI2000-77,KBSE85
Volume (vol) vol.100
Number (no) 709
Page pp.pp.-
#Pages 2
Date of Issue