Poorvaja2001/CTO

该数据集用于预测药物开发中的临床试验结果，是《Automatically Labeling Clinical Trial Outcomes: A Large-Scale Benchmark for Drug Development》工作中提出的部分。具体包括：human_labels包含手动标注的子集，遵循与自动标注步骤相同的规则以移除简单案例；linkage包含基于文本相似性预测的弱关联试验；news包含与试验最相似的10个新闻标题及其情感分析；phase1_CTO_preds、phase2_CTO_preds、phase3_CTO_preds包含各阶段特定的CTO预测，关键列包括试验ID、预测结果和成功概率；pubmed_gpt包含相关试验出版物的GPT预测；stocks_and_amendments包含股票价格斜率和试验修正次数。数据集基于CTTI的AACT数据库构建。Dataset for predicting clinical trial outcomes in drug development. This dataset is part of the work presented in Automatically Labeling Clinical Trial Outcomes: A Large-Scale Benchmark for Drug Development. Descriptions: human_labels contains the manually annotated subset. linkage contains the weakly linked trials as predicted based on text similarity. news contains the top 10 most similar headlines to the trial. phase1_CTO_preds, phase2_CTO_preds, phase3_CTO_preds contain phase-specific CTO predictions. pubmed_gpt contains linked GPT predictions on related trial publications. stocks_and_amendments contains scraped stock price Slope and number of amendments. Based on CTTIs AACT database.