File Exchange

image thumbnail

OnlineGradedPossibi​​listicClustering

version 1.3 (1.24 MB) by

Implementation of Online Graded Possibilistic Clustering "OGPC" Clustering in MATLAB

1 Download

Updated

View License

Implementation of Online Graded Possibilistic Clustering "OGPC" Clustering in MATLAB.
An algorithm that is able to detect outliers and adapt to concept shift drift in data streams.
Read more about the online algorithm "https://www.springerprofessional.de/graded-possibilistic-clustering-of-non-stationary-data-streams/12047532.
Notes:
- This a modified version of the online algorithm (Algorithm able to detect outliers and adapt to concept shift).
- The algorithm can work on raw (But you will have to change dynamic plot axis limit to visualize the clusters centroids) and on normalized data (Apply normalizer on the training set).
Developer: Amr Abdullatif (DIBRIS-University of Genoa)
Publisher: DIBRIS (www.dibris.unige.it)
Contact Info: amr.r.abdullatif@gmail.com
Function:
[rhovals,summembership,U, Youtn, Y, normvals, bend, uval, yval] = ogpc(X, Y, fb, K, maxiter, eta0, alphamin, plt)

Comments and Ratings (1)

Eueg Eueg

Updates

1.3

Change in
- learning region parameters (Concept drift "slow learn", outliers "no learn", Concept shift "fast learn").
- Cluster width update function.

1.2

Change
Update beta inside ogpc.m
" bi0 = bend ; "

1.1

Update description structure.

1.1

Update description.

1.1

Modify title and description.

MATLAB Release
MATLAB 9.2 (R2017a)

Download apps, toolboxes, and other File Exchange content using Add-On Explorer in MATLAB.

» Watch video

html/