Skip to content
This repository has been archived by the owner on Mar 7, 2022. It is now read-only.
/ wdbscan Public archive

Python implementation of a weighted extension of DBSCAN (Ester et al., 1996).

License

Notifications You must be signed in to change notification settings

lucboruta/wdbscan

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

wdbscan

Python implementation of a weighted extension of DBSCAN (Ester et al., 1996).

This module provides a weighted implementation of Ester et al.'s algorithm, accomodating instance weights. Here, a core object is not defined as an object whose epsilon-neighborhood contains at least a given number of objects, but as an object whose epsilon-neighborhood weighs a least a given weight. The weight of a given object's epsilon-neighborhood is straightforwardly defined as the sum of the weights of its epsilon-neighbors. This implementation does not require weights to be positive, so be careful what you wish for, you might just get it.

To use this module, you must have NumPy installed.

License

Do What The Fuck You Want To Public License (version 2).

Citation

When using wdbscan for a publication, please cite my doctoral dissertation.

@phdthesis{
    author = {Boruta, Luc},
    title = {Indicators of Allophony and Phonemehood},
    school = {Universit{\'e} Paris Diderot},
    year = {2012},
}

About

Python implementation of a weighted extension of DBSCAN (Ester et al., 1996).

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages