Research map / Time series / Causal and generative time series
Relation Inference among Sensor Time Series in Smart Buildings with Metric Learning
Shuheng Li, Dezhi Hong, Hongning Wang
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 |
|---|---|---|---|
| add STFT frequency-domain input (vs raw time series, TN) (ablation) | STN · Spatial relation inference | edge accuracy (%): 25.73 ± 1.94 -> 88.61 ± 2.08 | Improved |
| add STFT input (vs TN) on noisy building 10320 (ablation) | STN · Building 10320 VAV assignment | accuracy (%): 42.73 -> 90.00 | Improved |
| use triplet loss instead of Siamese (SSN) (ablation) | STN · Spatial relation inference | edge accuracy (%): 67.93 ± 8.66 -> 88.61 ± 2.08 | Improved |
| use triplet loss instead of Siamese on building 10320 (ablation) | STN · Building 10320 VAV assignment | accuracy (%): 42.73 -> 90.00 | Improved |
| use triplet loss instead of Siamese on building 10312 (ablation) | STN · Building 10312 VAV assignment | accuracy (%): 97.69 -> 98.07 | No clear effect |
| use indirectly correlated sensor pair (SpaceTemp/SupplyFanSpeed) instead of best pair | STN · Building VAV assignment | accuracy (%): 98.07 -> 96.05 | Mixed |
| train on one month instead of ten months | STN · Building 10320 VAV assignment | accuracy (%): 90.00 -> 47.27 | Hurt |
Closest papers by meaning
- A Scalable Multi-Task Model for Virtual Sensors (NeurIPS 2026 · similarity 0.83)
- Graph-based Virtual Sensing from Sparse and Partial Multivariate Observations (ICLR 2024 · similarity 0.81)
- Building Timeseries Dataset: Empowering Large-Scale Building Analytics (NeurIPS 2024 · similarity 0.81)
- Predicting User Behavior in Smart Spaces with LLM-Enhanced Logs and Personalized Prompts (AAAI 2025 · similarity 0.81)
- Real-Time Calibration Model for Low-Cost Sensor in Fine-Grained Time Series (AAAI 2025 · similarity 0.81)
- Fully-Connected Spatial-Temporal Graph for Multivariate Time-Series Data (AAAI 2024 · similarity 0.80)
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.
