🧪 Test?View on arXiv
A Large-Scale Dataset for Multi-Class Financial News Classification in Brazilian Portuguese
NLPfinancial newsclassificationBrazilian Portuguese
2610.20630
Builder Relevance
1h ago70%
Abstract
The paper presents a large-scale corpus of approximately 160,000 Brazilian Portuguese financial news articles to aid in the development of Financial NLP systems.
Reality Card
Core Claim
The study establishes foundational baselines for multi-class financial news classification in Brazilian Portuguese using a large-scale dataset and evaluates various similarity-based methods.
Method / Result
KNN achieved the best performance with 47.3% accuracy and 31.9% F1-score on the manually labeled subset.
Limitations
The task is highly sensitive to label quality and class distribution, with significant issues arising from label imbalance and noisy weak annotations.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.