Fuzzy logic c-means clustering
Fuzzy Clustering and Data Analysis
Find your system's clusters with two different fuzzy clustering algoritms (Fuzzy C-Means; Gustafson Kessel) and analyse with Least Square Estimation or Fuzzy Functions. You can analyse up to 8 (7 inputs, 1 output) dimensions.
How to use
- Load your data using Load Data window. After paste data into the text area, click Accept button. Application will review and accept max 2000 lines and 8 columns of your data. You can review what application has as data.
- Choose have may rows of data will be use as validation data. Train data is must. So you can choose max 40 lines of validation data.
- Choose how many center are in the data.
- Choose fuzziness coefficient. Best values are between 1.4 and 2.8 .
- Choose method to analyse.
- Click Calculate button.
- Get the results by Results button.
- Ported to aydos.com
- Calculation for one cluster added
- Small changes
- Min-max of graph
- Clustering method: Gustafson-Kessel
- User friendly graphic interface, tabs
- Graphs of data, clusters, memberships and estimates
- Visitor can analyse data of size 4x400; members can 6x1000
- Comparation current calculation with previous one
- Basic graphic interface
- Clustering method: FCM (Fuzzy c-mean clustering)
- Analysis method: Fuzzy Least Square Estimation
- Analysis method: Fuzzy Functions
- Prof.Dr. Lotfi A. Zadeh
- Prof.Dr. Burhan Türkşen
- Minimal Comps by Keith Peters