https://github.com/borgwardtlab/s3m
A software package for statistically significant shapelet mining
https://github.com/borgwardtlab/s3m
biomedical biomedical-data-science shapelets significant-pattern-mining time-series time-series-analysis time-series-analysis-tool
Last synced: 7 months ago
JSON representation
A software package for statistically significant shapelet mining
- Host: GitHub
- URL: https://github.com/borgwardtlab/s3m
- Owner: BorgwardtLab
- License: mit
- Created: 2018-03-05T14:22:17.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2022-12-02T15:24:32.000Z (over 2 years ago)
- Last Synced: 2023-10-20T18:23:27.011Z (over 1 year ago)
- Topics: biomedical, biomedical-data-science, shapelets, significant-pattern-mining, time-series, time-series-analysis, time-series-analysis-tool
- Language: C++
- Size: 2.88 MB
- Stars: 52
- Watchers: 6
- Forks: 8
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# S3M — Statistically Significant Shapelet Mining
This is the implementation of the method described in the ISMB 2018
paper [“Association mapping in biomedical time series via
statistically significant shapelet
mining”](https://doi.org/10.1093/bioinformatics/bty246).
[](https://travis-ci.com/BorgwardtLab/S3M)
# Installation
For Mac OS X, we recommend installing S3M using the [`Homebrew` package
manager](https://brew.sh):
$ brew install BorgwardtLab/mlcb/s3m
For Debian and Ubuntu, please use the `.deb` file of the [latest release](https://github.com/BorgwardtLab/S3M/releases/latest) and install it using:
$ sudo apt install s3m-latest.deb
For Arch Linux, please install S3M from [AUR](https://aur.archlinux.org/packages/s3m-git)
using [`pacaur`](https://github.com/rmarquis/pacaur) or
[`trizen`](https://github.com/trizen/trizen), for example:
$ pacaur -S s3m
$ trizen -S s3m
Please refer to the [advanced build instructions](docs/advanced_build.md)
for more advanced build processes, including source builds and Docker
installations.
# Example
We provide a synthetic data set that illustrates the utility of S3M.
After installation, the command
$ s3m -i data/example/synthetic.csv -m 15 -o results/example.json
runs S3M on it and stores its results in the `results` folder. The below
figure shows how the method identifies the case-characteristic two
spikes as the most significant shapelet (p-value: 5.42e-20).
# Help
If you have questions concerning S3M or you encounter problems when
trying to build the tool under your own system, please open an issue in
[the issue tracker](https://github.com/BorgwardtLab/S3M/issues). Try to
describe the issue in sufficient detail in order to make it possible for
us to help you.
# Contributors
S3M is developed and maintained by members of the [Machine Learning and
Computational Biology Lab](https://www.bsse.ethz.ch/mlcb) of [Prof. Dr.
Karsten Borgwardt](https://www.bsse.ethz.ch/mlcb/karsten.html):
- Christian Bock ([GitHub](https://github.com/chrisby))
- Thomas Gumbsch ([GitHub](https://github.com/tgumbsch))
- Max Horn ([GitHub](https://github.com/maexlich))
- Michael Moor ([GitHub](https://github.com/mi92))
- Bastian Rieck ([GitHub](https://github.com/Submanifold))
# Citation
Please use the following BibTeX citation when using our method or
comparing against it:
```TeX
@article{Bock18,
author = {Bock, Christian and Gumbsch, Thomas and Moor, Michael and Rieck, Bastian and Roqueiro, Damian and Borgwardt, Karsten},
title = {Association mapping in biomedical time series via statistically significant shapelet mining},
journal = {Bioinformatics},
volume = {34},
number = {13},
pages = {i438--i446},
year = {2018},
doi = {10.1093/bioinformatics/bty246},
}
```