Matching hierarchical structures, like taxonomies or web directories, is the premise for enabling interoperability among heterogenous data organizations. While the number of new matching solutions is increasing the evaluation issue is still open. This work addresses the problem of comparison for pairwise matching solutions. A methodology is proposed to overcome the issue of scalability. A large scale dataset is developed based on real world case study namely, the web directories of Google, Looksmart and Yahoo!. Finally, an empirical evaluation is performed which compares the most representative solutions for taxonomy matching. We argue that the proposed dataset can play a key role in supporting the empirical analysis for the research effort in the area of taxonomy matching.
A Large Scale Taxonomy Mapping Evaluation / Avesani, Paolo; Giunchiglia, Fausto; Yatskevich, Mikalai. - ELETTRONICO. - (2005), pp. 1-17.
A Large Scale Taxonomy Mapping Evaluation
Avesani, Paolo;Giunchiglia, Fausto;Yatskevich, Mikalai
2005-01-01
Abstract
Matching hierarchical structures, like taxonomies or web directories, is the premise for enabling interoperability among heterogenous data organizations. While the number of new matching solutions is increasing the evaluation issue is still open. This work addresses the problem of comparison for pairwise matching solutions. A methodology is proposed to overcome the issue of scalability. A large scale dataset is developed based on real world case study namely, the web directories of Google, Looksmart and Yahoo!. Finally, an empirical evaluation is performed which compares the most representative solutions for taxonomy matching. We argue that the proposed dataset can play a key role in supporting the empirical analysis for the research effort in the area of taxonomy matching.File | Dimensione | Formato | |
---|---|---|---|
044.pdf
accesso aperto
Tipologia:
Versione editoriale (Publisher’s layout)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
211.93 kB
Formato
Adobe PDF
|
211.93 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione