← Back to all news
arXiv Machine Learning By Selim F. Yilmaz, E. Batuhan Kaynak, Aykut Ko\c{c}, Hamdi Dibeklio\u{g}lu, Suleyman S. Kozat

Multi-Label Sentiment Analysis on 100 Languages with Dynamic Weighting for Label Imbalance

Read the original on arXiv Machine Learning →

The paper presents a multi‑label sentiment analysis framework that follows Plutchik’s wheel of emotions and is applicable to 100 languages. It introduces a dynamic weighting scheme to balance class contributions during training, adapts focal loss for harder instances, and derives a linear‑time method for selecting optimal class‑specific thresholds to maximize macro‑F1. Experiments show state‑of‑the‑art performance across multiple languages and metrics, and the authors release the code publicly.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Computation and Language
2d ago

Steering with Multilingual Lexicons for Sentiment Control in Low-Resource Languages

The paper proposes using multilingual sentiment lexicons to steer language models for sentiment control, especially in low-resource languages. It finds that aggregated multilingual lexicons outperform English-derived steering vectors for mid- and low-resource languages, while monolingual lexicons are still best for languages the model was pre-trained on. The study also explains how representational density, pre-training priors, and architectural biases affect steering effectiveness, noting that large English vectors can cause linguistic collapse in low-resource settings but this is mitigated by norm-dampening in multilingual aggregation.

By Lucinda Laurent, Muhammad Dehan Al Kautsar, Alfan Farizki Wicaksono, Fajri Koto
More like this →
arXiv Computation and Language
2d ago

Sentiment sensing: comparative analysis of multilingual sentiment dynamics

The paper introduces "sentiment sensing," a framework designed to overcome key obstacles in cross‑country sentiment analysis, such as language‑specific biases, compositional data constraints, and the lack of rigorous testing for temporal dynamics. It integrates compositional data analysis, functional data methods, and non‑parametric inference to allow valid multilingual comparisons and statistically defensible detection of sentiment shifts. The authors demonstrate the framework in two case studies, analyzing news articles from all 27 EU member states on energy and artificial intelligence topics.

By Mara Sabina Bernardi, Elisa Di Nuovo, Francesca Torti, Giuseppe Tirone, Nicolas Stefanovitch, Zsuzsanna Pasztor, Charles Macmillan, Domenico Perrotta, Bertrand De Longueville
nlp
More like this →