Reproduce “Financial Sentiment Analysis and Classification: A Comparative Study of Fine-Tuned Deep Learning Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Dimitrios K. Nasiopoulos, Konstantinos I. Roumeliotis, Damianos P. Sakas, Kanellos Toudas, Panagiotis Reklitis
The authors compare pretrained GPT-4o-mini models before and after fine-tuning for three-way classification of financial sentences as negative, positive, or neutral. On one balanced test split of 516 sentences, fine-tuning raises GPT-4o-mini's accuracy from 0.7752 to 0.8779, matching fine-tuned GPT-4o. The result concerns sentiment labeling, not market prediction or performance on other financial text sources.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.