Research map / Graph learning / Knowledge and temporal graphs
Temporal Positive-unlabeled Learning for Biomedical Hypothesis Generation via Risk Estimation
Uchenna Akujuobi, Jun Chen, Mohamed Elhoseiny, Michael Spranger, Xiangliang Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use temporal positive-unlabeled risk estimation (TRP-UPU) rather than the PN variant that labels unobserved pairs negative (ablation) | TRP · COVID-19 | F1-S: 0.84 -> 0.86 | Improved |
| Use temporal positive-unlabeled risk estimation rather than the PN variant (ablation) | TRP · Virology | F1-S: 0.73 -> 0.74 | Improved |
| Use temporal positive-unlabeled risk estimation rather than the PN variant (ablation) | TRP · Immunotherapy | F1-S: 0.71 -> 0.71 | No clear effect |
| Use unbiased PU risk estimation instead of non-negative PU risk estimation (ablation) | TRP · COVID-19 | F1-S: 0.80 -> 0.86 | Improved |
| Use unbiased PU risk estimation instead of non-negative PU risk estimation (ablation) | TRP · Virology | F1-S: 0.73 -> 0.74 | Improved |
| Use unbiased PU risk estimation instead of non-negative PU risk estimation (ablation) | TRP · Immunotherapy | F1-S: 0.67 -> 0.71 | Improved |
| Use temporal modeling with the TRP-UPU framework rather than the strongest reported non-TRP baseline | Biomedical term link prediction · COVID-19 | F1-S: 0.85 -> 0.86 | Improved |
| Use temporal modeling with the TRP-UPU framework rather than the strongest reported non-TRP baseline | Biomedical term link prediction · Virology | F1-P: 2.45 -> 2.38 | Hurt |
| Use temporal modeling with the TRP-UPU framework rather than the strongest reported non-TRP baseline | Biomedical term link prediction · Immunotherapy | F1-P: 2.26 -> 2.35 | Improved |
Closest papers by meaning
- Causal Hidden Markov Model for Time Series Disease Forecasting (CVPR 2021 · similarity 0.84)
- Context-Aware Health Event Prediction via Transition Functions on Dynamic Disease Graphs (AAAI 2022 · similarity 0.83)
- Risk Horizons: Structured Hypothesis Spaces for Longitudinal Clinical Prediction (NeurIPS 2026 · similarity 0.83)
- Generating Hypotheses of Dynamic Causal Graphs in Neuroscience: Leveraging Generative Factor Models of Observed Time Series (ICML 2025 · similarity 0.83)
- Towards Enhancing Time Series Contrastive Learning: A Dynamic Bad Pair Mining Approach (ICLR 2024 · similarity 0.82)
- JoLT: Jointly Learned Representations of Language and Time-Series for Clinical Time-Series Interpretation (Student Abstract) (AAAI 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
