Lexico-Semantic Norms¶
TRUNAJOD lexico semantic norms module.
Lexico-Semantic norms do also require external knowledge to be computed. We compute the following lexico-semantic variables:
Arousal
Concreteness
Context Availability
Familiarity
Imageability
Valence
We provide two downloadable models of these variables, which come from [DPSebastianGalles+13] and [GFerreF16].
-
class
TRUNAJOD.lexico_semantic_norms.LexicoSemanticNorm(doc, lexico_semantic_norm_dict, lemmatizer=None)¶ Create a lexico semantic norm calculator for text.
This requires a lexico semantic norm dict, with key-value pairs specified as
word -> {"arousal", "concreteness", "context_availability", "familiarity", "imageability", "valence"}. Average over number of tokens will be computed. The values are obtained from [GFerreF16].-
get_arousal()¶ Get arousal.
- Returns
Average arousal.
- Return type
float
-
get_concreteness()¶ Get concreteness.
- Returns
Average concreteness.
- Return type
float
-
get_context_availability()¶ Get context_availability.
- Returns
Average context_availability.
- Return type
float
-
get_familiarity()¶ Get familiarity.
- Returns
Average familiarity.
- Return type
float
-
get_imageability()¶ Get imageability.
- Returns
Average imageability.
- Return type
float
-
get_valence()¶ Get valence.
- Returns
Average valence.
- Return type
float
-
-
TRUNAJOD.lexico_semantic_norms.get_conc_imag_familiarity(doc)¶ Get lexico-semantic variables.
Computes three lexico-semantic variables: Concreteness, Imageability and Familiarity. The values are obtained from the EsPal dictionary (Spanish) and average of each metric is computed over sentences. To get each metric, the best practice is using LSNorm Enum defined in lexicosemantic_norms_espal module. The enums are CONCRETENESS, IMAGEABILITY and FAMILIARITY. This implementation uses values of the lexico-semantic norms from [DPSebastianGalles+13].
- Parameters
doc (Spacy Doc) – Tokenized text
- Returns
Concreteness imageability and familiarity averaged over sentences
- Return type
List of float
- DPSebastianGalles+13(1,2)
Andrew Duchon, Manuel Perea, Nuria Sebastián-Gallés, Antonia Martí, and Manuel Carreiras. Espal: one-stop shopping for spanish word properties. Behavior research methods, 45(4):1246–1258, 2013.
- GFerreF16(1,2)
Marc Guasch, Pilar Ferré, and Isabel Fraga. Spanish norms for affective and lexico-semantic variables for 1,400 words. Behavior Research Methods, 48(4):1358–1369, 2016.