Searching for services often starts from the exploration ofthe service space. Community generated tags can support exploration of this space. Researchers attracted by the community-available “free manpower” proposed more complex tagging-based annotation models.Those models tag specific parts of the Web service definition: single operations, as well as inputs and outputs of those operations. However, there is no evidence whether effort spent on annotating complex structures is justified by the performance improvement of retrieval based on those annotations. In this paper we apply similarity-based search to find a trade-off between retrieval effectiveness of existing annotation modelsand the annotation effort.
Trade-off between complexity of structured tagging and effectiveness of Web service retrieval / Gawinecki, Maciej; Cabri, Giacomo; M., Paprzycki; M., Ganzha. - STAMPA. - 6385:(2010), pp. 289-300. (Intervento presentato al convegno 10th International Conference on Web Engineering, ICWE 2010 tenutosi a Vienna, aut nel July 2010) [10.1007/978-3-642-16985-4_26].
Trade-off between complexity of structured tagging and effectiveness of Web service retrieval
GAWINECKI, MacieJ;CABRI, Giacomo;
2010
Abstract
Searching for services often starts from the exploration ofthe service space. Community generated tags can support exploration of this space. Researchers attracted by the community-available “free manpower” proposed more complex tagging-based annotation models.Those models tag specific parts of the Web service definition: single operations, as well as inputs and outputs of those operations. However, there is no evidence whether effort spent on annotating complex structures is justified by the performance improvement of retrieval based on those annotations. In this paper we apply similarity-based search to find a trade-off between retrieval effectiveness of existing annotation modelsand the annotation effort.
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris
