用结构化特征预测创始人成功，发现数据信号有限且已到瓶颈。
When Career Data Runs Out: Structured Feature Engineering and Signal Limits for Founder Success Prediction
- 从职业履历直接提取28个结构化特征，结合规则层与梯度提升模型
- 模型在验证集上F0.5达0.303，比零样本大模型提升17.7个百分点
- 大模型从文字中提取特征无效，因原始数据已被损失性重编码
从创始人职业数据预测初创企业成功极为困难：信号微弱，标签稀少（仅9%），多数成功与失败者几乎相同。我们从原始JSON字段（工作、教育、退出经历）中直接构建28个结构化特征，结合确定性规则层与XGBoost增强的决策树桩。模型在验证集上取得F0.5 = 0.3030，精确率 = 0.3333，召回率 = 0.2222，较零样本大模型基线提升17.7个百分点。随后进行控制实验：使用Claude Haiku从非结构化文本中提取9个特征，在67%和100%数据覆盖下，这些特征虽占模型重要性26.4%，但未带来任何交叉验证性能提升（Δ = -0.05pp）。原因是匿名化文本由同一组原始数据生成，为有损重编码，非信息增益源。模型上限（交叉验证≈0.25，验证≈0.30）反映该数据集的信息容量，而非建模能力。本研究作为诊断基准，明确指出信号耗尽位置与成因，直接指明更优数据集应包含的内容。
原文摘要 · Abstract (English)
Predicting startup success from founder career data is hard. The signal is weak, the labels are rare (9%), and most founders who succeed look almost identical to those who fail. We engineer 28 structured features directly from raw JSON fields -- jobs, education, exits -- and combine them with a deterministic rule layer and XGBoost boosted stumps. Our model achieves Val F0.5 = 0.3030, Precision = 0.3333, Recall = 0.2222 -- a +17.7pp improvement over the zero-shot LLM baseline. We then run a controlled experiment: extract 9 features from the prose field using Claude Haiku, at 67% and 100% dataset coverage. LLM features capture 26.4% of model importance but add zero CV signal (delta = -0.05pp). The reason is structural: anonymised_prose is generated from the same JSON fields we parse directly -- it is a lossy re-encoding, not a richer source. The ceiling (CV ~= 0.25, Val ~= 0.30) reflects the information content of this dataset, not a modeling limitation. In characterizing where the signal runs out and why, this work functions as a benchmark diagnostic -- one that points directly to what a richer dataset would need to include.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。