LEIT-motifs: Scalable Motif Mining in Multidimensional Time Series
Matteo Ceccarello, Francesco Pio Monaco, Francesco Silvestri · 2025
Read LEIT-motifs: Scalable Motif Mining in Multidimensional Time Series by Matteo Ceccarello, Francesco Pio Monaco, Francesco Silvestri (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-14446.txt.
Description
Ceccarello, Monaco, and Silvestri speed up the search for repeating patterns in data streams that carry many signals at once. Such patterns, called motifs, matter in fields from health monitors to finance. Their 2025 method, named MOMENTI in the text, finds the top patterns far faster than exact methods, whose run time grows with the square of the stream length.
The method promises the exact answer with a high chance set by the user. It tunes itself to the shape of the incoming data while respecting a memory cap the user sets. In wide tests it ran orders of magnitude faster than the best prior methods. The authors back the design with a full math analysis of its speed and memory use.
Catalog record
| CREATOR | Matteo Ceccarello, Francesco Pio Monaco, Francesco Silvestri |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-14446