Building on previous work on Weighted Description Logic (WDL), we present and assess an algorithm for concept combination grounded in the experimental research in cognitive psychology. Starting from two WDL formulas representing concepts in a way similar to Prototype Theory and a knowledge base (KB) modelling background knowledge, the algorithm outputs a new WDL formula which represent the combination of the input concepts. First, we study the logical properties of the operator defined by our algorithm. Second, we collect data on the prototypical representation of concepts and their combinations and learn WDL formulas from them.Third, we evaluate our algorithm and the role of the KB by comparing the algorithm's outputs with the learned WDL formulas.
Concept Combination in Weighted DL
Masolo, Claudio
2023
Abstract
Building on previous work on Weighted Description Logic (WDL), we present and assess an algorithm for concept combination grounded in the experimental research in cognitive psychology. Starting from two WDL formulas representing concepts in a way similar to Prototype Theory and a knowledge base (KB) modelling background knowledge, the algorithm outputs a new WDL formula which represent the combination of the input concepts. First, we study the logical properties of the operator defined by our algorithm. Second, we collect data on the prototypical representation of concepts and their combinations and learn WDL formulas from them.Third, we evaluate our algorithm and the role of the KB by comparing the algorithm's outputs with the learned WDL formulas.| File | Dimensione | Formato | |
|---|---|---|---|
|
Righetti2023-JELIA_concept_combination.pdf
solo utenti autorizzati
Descrizione: Righetti, G., Galliani, P., Masolo, C. (2023). Concept Combination in Weighted DL. In: Gaggl, S., Martinez, M.V., Ortiz, M. (eds) Logics in Artificial Intelligence. JELIA 2023. Lecture Notes in Computer Science(), vol 14281. Springer, Cham. https://doi.org/10.1007/978-3-031-43619-2_27
Tipologia:
Versione Editoriale (PDF)
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
373.94 kB
Formato
Adobe PDF
|
373.94 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.
