knowtrendllc/earnings_call_transcript_autograder

该数据集包含对约70,000+份财报电话会议记录的LLM生成分析。分析使用了Kimi k2-0905-preview模型，专注于提取特定见解、有先见之明的分析师问题和管理失误。数据以Parquet格式存储，具有扁平化的列式架构，便于转换为Pandas、SQL或直接在Hugging Face查看器中过滤。数据集包括元数据和财务数据、评分（1-10）、回报以及奖项（如“侦探奖”、“小丑奖”、“诺查丹玛斯奖”和“大嘴巴奖”），每个奖项都有详细的列描述。

This dataset contains LLM-generated analysis of ~70,000+ earnings call transcripts. The analysis was performed using Kimi k2-0905-preview, focusing on extracting specific insights, prescient analyst questions, and management missteps. The data is stored in Parquet format with a flattened, columnar schema, making it easy to convert to Pandas, SQL, or filter directly in the Hugging Face viewer. The dataset includes metadata & financials, scores (1-10), returns, and awards (e.g., detective_, joker_, nostradamus_, loudmouth_), each with detailed column descriptions.