https://github.com/active-group/active-analytics
https://github.com/active-group/active-analytics
Last synced: 6 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/active-group/active-analytics
- Owner: active-group
- License: epl-1.0
- Created: 2018-12-04T11:32:13.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2019-01-22T08:24:11.000Z (over 7 years ago)
- Last Synced: 2024-11-29T01:49:28.444Z (over 1 year ago)
- Language: Clojure
- Size: 68.4 KB
- Stars: 0
- Watchers: 15
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Active Analytics
A Clojure library containing analytic methods used in our projects.
#### Dependency information
[Leiningen](https://github.com/technomancy/leiningen/)
[de.active-group/active-analytics "0.3.0"]
[deps.edn](https://clojure.org/guides/deps_and_cli)
{de.active-group/active-analytics {:mvn/version "0.3.0"}}
## Usage
So far, there are three implemented clustering algorithms:
- [Power iteration clustering (PIC)](http://www.cs.cmu.edu/~frank/papers/icml2010-pic-final.pdf): `active-analytics.clustering.pic`
- [k-means clustering](https://en.wikipedia.org/wiki/K-means_clustering): `active-analytics.clustering.k-means`
- [k-medoids clustering](https://en.wikipedia.org/wiki/K-medoids) (Lloyd's algorithm): `active-analytics.clustering.k-medoids`
At the moment, the PIC implementation as well as the default k-means
centroid function only support [neanderthal](https://neanderthal.uncomplicate.org/)
as backend. Please follow its [instructions](https://neanderthal.uncomplicate.org/articles/getting_started.html#the-native-library-used-by-neanderthals-native-engine)
on how to install the needed native Intel MKL library.
The `active-analytics.clustering.silhouettes` namespace contains functions to
measure the quality of clustering results.
## License
Copyright © 2018 Active Group GmbH
Distributed under the Eclipse Public License either version 1.0 or (at
your option) any later version.