Time2State: An Unsupervised Framework for Inferring the Latent States in Time Series Data
Summary: Unsupervised Time2State infers latent states in massive multivariate time series, exposing high-level semantics (e.g., run, walk, jump). A sliding-window encoder with a novel LSE-Loss reduces computational cost and yields up to 15% accuracy gains over prior time-series representation methods. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Chengyu Wang
- 2. Kui Wu
- 3. Tongqing Zhou
- 4. Zhiping Cai
Incoming Citations (Sorted by Pagerank)
Showing 2 of 2 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 9,147 | ISSD: Indicator Selection for Time Series State Detection | 2025 | SIGMOD | 4.3849295e-05 |
| 10,309 | CLaP - State Detection from Time Series | 2026 | VLDB | 4.1945683e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 2 of 2 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 270 | OPTICS: Ordering Points To Identify the Clustering Structure | 1999 | SIGMOD | 0.00029505642 |
| 4,065 | AutoPlait: Automatic Mining of Co-evolving Time Sequences | 2014 | SIGMOD | 6.4819215e-05 |
Previous
Page 1 / 1
Next