Expert Finding (EF) techniques help in discovering people having relevant knowledge and skills. But for their validation, EF techniques usually rely on experts, meaning using another EF technique, generally not properly validated, and exploit them mainly for output validations, meaning only at late stages. We propose a model, which builds on literature in Psychology and practice, to identify generic concepts and relations in order to support the analysis and design of EF techniques, thus inferring potential improvements during early stages in an expertfree manner. Our contribution lies in the identification and review of relevant literature, building the conceptual model, and illustrating its use through an analysis of existing EF techniques. Although the model can be improved, we can already identify strengths and limitations in recent EF techniques, thus supporting the usefulness of a model-based analysis and design for EF techniques.
Breaking the Recursivity: Towards a Model to Analyse Expert Finders
Vergne, Matthieu Dominique Patrick;Susi, Angelo
2015-01-01
Abstract
Expert Finding (EF) techniques help in discovering people having relevant knowledge and skills. But for their validation, EF techniques usually rely on experts, meaning using another EF technique, generally not properly validated, and exploit them mainly for output validations, meaning only at late stages. We propose a model, which builds on literature in Psychology and practice, to identify generic concepts and relations in order to support the analysis and design of EF techniques, thus inferring potential improvements during early stages in an expertfree manner. Our contribution lies in the identification and review of relevant literature, building the conceptual model, and illustrating its use through an analysis of existing EF techniques. Although the model can be improved, we can already identify strengths and limitations in recent EF techniques, thus supporting the usefulness of a model-based analysis and design for EF techniques.File | Dimensione | Formato | |
---|---|---|---|
main.pdf
solo utenti autorizzati
Descrizione: Main article
Tipologia:
Documento in Pre-print
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
218.45 kB
Formato
Adobe PDF
|
218.45 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.