Solving Morphological Analogies Through Generation

Kevin Chan (1) , Shane Peter Kaszefski-Yaschuk , Camille Saran (1) , Esteban Marquer (2) , Miguel Couceiro (2)

1 IDMC - Institut des Sciences du Digital, Management et Cognition
2 ORPAILLEUR - Knowledge representation, reasonning

This contribution is a first attempt at solving morphological analogies through generation, instead of relying on retrieval approaches. Our preliminary experiments show promising results for some languages and reveal the feasibility of the approach in generating solutions of analogical equations in the morphology setting.

Mots clés: Analogy solving; Representation learning; Word generation; Morphological analogy

Domaines: Intelligence artificielle [cs.AI]; Apprentissage [cs.LG]

Kevin Chan, Shane Peter Kaszefski-Yaschuk, Camille Saran, Esteban Marquer, Miguel Couceiro. Solving Morphological Analogies Through Generation. IJCAI-ECAI Workshop on the Interactions between Analogical Reasoning and Machine Learning (IARML@IJCAI-ECAI 2022), Miguel Couceiro and Pierre-Alexandre Murena, Jul 2022, Vienna, Austria. pp.29-39. ⟨hal-03674913⟩