Conducting term alignment of a dataset without data provider identification

Authors:
Tetsuya Yoshida
Affiliations:
Hokkaido University, Graduate School of Information Science and Technology, Sapporo, Hokkaido, Japan
Venue:
ICCOMP'06 Proceedings of the 10th WSEAS international conference on Computers
Year:
2006

Citing 8
Cited 0

A translation approach to portable ontology specifications

Knowledge Acquisition - Special issue: Current issues in knowledge modeling
Advances in knowledge discovery and data mining

Advances in knowledge discovery and data mining
Readings in GroupWare and Computer-Supported Cooperative Work: Assisting Human-Human Collaboration

Readings in GroupWare and Computer-Supported Cooperative Work: Assisting Human-Human Collaboration
Induction of Decision Trees

Machine Learning
OBSERVER: An Approach for Query Processing in Global Information Systems based on Interoperation across Pre-existing Ontologies

COOPIS '96 Proceedings of the First IFCIS International Conference on Cooperative Information Systems
Automatic identification of word translations from unrelated English and German corpora

ACL '99 Proceedings of the 37th annual meeting of the Association for Computational Linguistics on Computational Linguistics
Constructing a Decision Tree for Graph-Structured Data and its Applications

Fundamenta Informaticae - Advances in Mining Graphs, Trees and Sequences
Detecting difference of usage of terms as difference of structure

Cognitive Systems Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes to extend our previous approach for term alignment of datasets so that term alignment can be conducted even when the identification of data provider is not known for each data item. Our previous method can conduct term alignment w.r.t. the usage of terms in data items, provided the identification of data provider is known for each data item. However, this assumption may not always hold, since data can be collected and uploaded from anonymous providers. To tackle this problem, this paper proposes a new method which partitions a dataset into subsets of data items. To seek for a better partition, decision trees are constructed for the subsets. By defining a distance between decision trees, the quality of a partition is measured in terms of the stability of the structure of the constructed decision trees w.r.t. the defined distance. Our previous method for term alignment is then applied for the partitioned subsets to conduct term alignment. The implementation of the proposed method has been conducted and its effectiveness is evaluated through experiments with new evaluation measures in the context of our term alignment.