An instance-based schema matching method with attributes ranking and classification

Authors:
Ji Feng;Xiaoguang Hong;Yuanbo Qu
Affiliations:
School of Computer Science and Technology, Shandong University, China;School of Computer Science and Technology, Shandong University, China;School of Computer Science and Technology, Shandong University, China
Venue:
FSKD'09 Proceedings of the 6th international conference on Fuzzy systems and knowledge discovery - Volume 5
Year:
2009

Citing 6
Cited 0

Generic Schema Matching with Cupid

Proceedings of the 27th International Conference on Very Large Data Bases
Database Schema Matching Using Machine Learning with Feature Selection

CAiSE '02 Proceedings of the 14th International Conference on Advanced Information Systems Engineering
Automatic direct and indirect schema mapping: experiences and lessons learned

ACM SIGMOD Record
COMA: a system for flexible combination of schema matching approaches

VLDB '02 Proceedings of the 28th international conference on Very Large Data Bases
Instance-based schema matching for web databases by domain-specific query probing

VLDB '04 Proceedings of the Thirtieth international conference on Very large data bases - Volume 30
Random Forests for multiclass classification: Random MultiNomial Logit

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Schema matching is a critical problem in many applications of database system, such as information integration, data warehouses, e-commerce, etc. So far, many solutions based on schema and element have been proposed. In this paper we present a new approach of instance-based matching building on the hypothesis that the corresponding attributes have equal relative importance. The framework of our apporach consists of three parts: attribute ranking, attribute classification and matching phase. Unlike traditional approaches considering all atrributes with the same importance, we take machine learning methods to prioritize all schema attributes by ranking and classification. During the matching phase, we construct an optimal objective function to find all equivalent attributes. In the end, our approach is validated by real datasets and the results show good accuarcy.