emolex - v1.0.0

Title:

• EmoLex: A Random Forest classifier for emotional lexica in Spanish

Creator:

• Name: Cárdenas-Mancilla, M. H.
• ORCID: https://orcid.org/0000-0002-6942-6232

Description:

This release contains the first version of a supervised classification model for emotional lexical profiling in Spanish. The Random Forest classifier predicts affective-semantic subgroups based on psycholinguistic variables (valence, arousal, concreteness, emotionality, frequency) and a novel metric, the Balanced Integration Score (BIS), derived from unsupervised clustering analyses. The model is trained on the EmoPro corpus (Perez-Sánchez et al., 2021) and outputs high-precision results using interpretable features. It includes full documentation in English and Spanish, and exports trained models with embedded timestamps for replication and deployment.

Keywords:

• lexical semantics
• affective processing
• supervised learning
• Random Forest
• BIS
• Spanish emotional lexicon
• machine learning
• psycholinguistics

license:

• AGPL-3.0-or-later

version: 1.0.0

publication_date:

• 2025-07-25

language:

• Spanish

related_identifiers:

• identifier: https://github.com/marcoscardenasmancilla/EmoLex_Classifier
• relation: isSupplementTo
• scheme: url

access_right:

• open

upload_type:

• software

Notes:

This model is designed for academic and research use. It complements prior work on the EmoPro corpus and the implementation of the Balanced Integration Score (BIS) for clustering-based lexical analysis. For citation, please refer to the associated publication or cite this release via DOI.

References:

• Pérez-Sánchez, M. Á., Stadthagen-Gonzalez, H., Guasch, M., Hinojosa, J. A., Fraga, I., Marín, J., & Ferré, P. (2021). EmoPro: Emotional prototypicality for 1,286 Spanish words: Relationships with affective and psycholinguistic variables. Behavior Research Methods, 53(5), 1857–1875. https://doi.org/10.3758/s13428-020-01519-9

- Python
Published by marcoscardenasmancilla 7 months ago