Abstract
Dictionaries are sociocultural objects that can be used as underlying structures of cognitive science models. We first show that the lexical networks constructed from dictionaries, despite a surface disagreement at links level, share a common topological structure. We assume that this deep structure reflects the semantic organisation of the lexicon shared by the members of a linguistic community. We propose a model based on the exploration of this specific struc- ture to analyse and compare the semantic efficiency of [Children/Adults] productions in action labelling tasks. We define a generic score of semantic efficiency, S KILLEX. Assigned to partic- ipants of the A PPROX protocol, this score enables us to accurately classify them into Children and Adults categories. Copyright © 2016 HAL.
| Original language | English |
|---|---|
| Pages (from-to) | 97-121 |
| Journal | TAL Traitement Automatique des Langues |
| Volume | 55 |
| Issue number | 3 |
| Publication status | Published - 2016 |
Citation
Gaume, B., Duvignau, K., Navarro, E., Desalle, Y., Cheung, H., Hsieh, S., Magistry, P., & Prévot, L. (2016). Skillex: a graph-based lexical score for measuring the semantic effi ciency of used verbs by human subjects describing actions. Traitement Automatique des Langues, 55(3), 97-121.Keywords
- Dictionaries
- Networks
- Lexicon
- Approx
- Skillex