Adaptive name matching in information integration

被引：202

作者：

Bilenko, M

Mooney, R

Cohen, W

Ravikumar, P

Fienberg, S

机构：

[1] Univ Texas, Dept Comp Sci, Austin, TX 78712 USA

[2] Carnegie Mellon Univ, Ctr Automated Learning & Discovery, Pittsburgh, PA 15213 USA

[3] Carnegie Mellon Univ, Sch Comp Sci, Pittsburgh, PA 15213 USA

[4] Carnegie Mellon Univ, Dept Stat, Pittsburgh, PA 15213 USA

[5] Carnegie Mellon Univ, Ctr Comp & Commun Secur, Pittsburgh, PA 15213 USA

来源：

IEEE INTELLIGENT SYSTEMS | 2003年 / 18卷 / 05期

基金：

美国安德鲁·梅隆基金会; 美国国家科学基金会;

关键词：

D O I：

10.1109/MIS.2003.1234765

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Different approaches to the name-matching problem in information integration are discussed. Those methods are used that adapt to a specific domain by combining multiple string similarity methods that capture different notions of similarity. Edit distance metrics are widely used and many variations are possible. An adaptive version of edit distance with affine gaps is proposed.

引用

页码：16 / 23

页数：8

共 27 条

[1] [Anonymous], 1999, INTERNAL REVENUE SER
[2] [Anonymous], P SIGMOD WORKSH DAT
[3] Bilenko M., 2003, Proc. 9th Int. Conf. Knowledge Discovery and Data Mining, Washington, P39, DOI DOI 10.1145/956750.956759
[4] CHURCHES T, 2002, MED INFORMATICS DEC, V2
[5] Cohen W. W., 2000, Proceedings. KDD-2000. Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, P255, DOI 10.1145/347090.347141
[6] Data integration using similarity joins and a word-based information representation language
Cohen, WW
[J]. ACM TRANSACTIONS ON INFORMATION SYSTEMS, 2000, 18 (03) : 288 - 321
[7] COHEN WW, 1998, P 1998 ACM SIGMOD IN, P201
[8] Durban R., 1998, BIOL SEQUENCE ANAL P
[9] A THEORY FOR RECORD LINKAGE
FELLEGI, IP
SUNTER, AB
[J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1969, 64 (328) : 1183 - &
[10] Galhardas H., 2001, Proceedings of the 27th International Conference on Very Large Data Bases, P371

← 1 2 3 →