List your company

Arabic Financial Sentiment Classification on Arabic Financial Sentiment Gold Standard (250 Samples)

61.6AccuracyDeepSeek R1 (Chat)
49.1252.3655.658.84May 19, 2026

Evaluation results

MethodLinks
DeepSeek R1 (Chat)2026.0561.647.361.653.436
GPT-52026.0557.260.257.258.382.9
DeepSeek R1 (Reasoner)2026.0556.857.156.855.673.9
GPT-4o Miniused to construct gold standard=true2026.0549.663.449.655.60
GPT-4 Turboused to construct gold standard=true2026.0549.663.449.655.60
Gemini 2.5 Flashused to construct gold standard=true2026.0549.663.449.655.60