Towards Spatial Word Embeddings - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Towards Spatial Word Embeddings

Résumé

Leveraging textual and spatial data provided in spatio-textual objects (eg., tweets), has become increasingly important in real-world applications, favoured by the increasing rate of their availability these last decades (eg., through smartphones). In this paper, we propose a spatial retrofitting method of word embeddings that could reveal the localised similarity of word pairs as well as the diversity of their localised meanings. Experiments based on the semantic location prediction task show that our method achieves significant improvement over strong baselines.
Fichier principal
Vignette du fichier
mousset_24752.pdf (458.53 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02494099 , version 1 (28-02-2020)

Identifiants

Citer

Paul Mousset, Yoann Pitarch, Lynda Tamine. Towards Spatial Word Embeddings. 41st European Conference on Information Retrieval (ECIR 2019), Apr 2019, Cologne, Germany. pp.53-61, ⟨10.1007/978-3-030-15719-7_7⟩. ⟨hal-02494099⟩
40 Consultations
145 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More