Hope Speech Detection: A Comparative Study across Four Languages

International Journal of Combinatorial Optimization Problems and Informatics. 2026 Open access

Read full text
Authors: Oluwatobi Joseph Abiola, Tolulope Olalekan Abiola, Obdulia Pichardo Lagunas, Grigori Sidorov
DOI: 10.61467/2007.1558.2026.v17i2.1279
Cited by: 0

Abstract

This paper presents a multilingual approach to hope speech detection using both traditional machine learning (Ridge Classifier with TF-IDF) and transformer-based models (LaBSE). We evaluate performance across four languages English, Spanish, Urdu, and German using the PolyHope dataset. Experimental results demonstrate that LaBSE consistently outperforms the Ridge baseline, particularly in low-resource settings like Urdu. Our findings highlight the effectiveness of multilingual transformers in capturing nuanced expressions of hope across diverse linguistic and cultural contexts. Smart citations: https://scite.ai/reports/10.61467/2007.1558.2026.v17i2.1279Dimensions.Open Alex.

Fields of Study & Subjects

Sentiment Analysis and Opinion Mining Mental Health via Writing Topic Modeling

Publication details

Journal / source
International Journal of Combinatorial Optimization Problems and Informatics.
Volume
17
Issue
2
Pages
215–221
Publication date
16 February 2026
Language
EN