Unsupervised Machine Learning Simulator - Download

a standalone windows application to cluster data

Posted on January 23, 2018

An easy-to-use windows application to simulate the two major Clustering algorithms for unsupervised machine learning, KMeans and Hierarchical Clustering. I'll keep updating versions to make the application more intelligent and usable with different kinds of datasets.

Clustering Simulator

The Simulator is a standalone app that does not require python or any other associated dependencies to be installed in your system.

  • Download the Simulator by clicking on the google drive link. Unsupervised ML Simulator v1.0 (Windows only)
  • Extract the contents of the zip file
  • You'll find two folders, input and output. The input folder is where you put your csv file that needs to be clustered.
  • To test the simulator with the default files just run the simulator application and you'll have your output files generated in the output folder.
Guidelines for input.csv and program settings.txt
  • Place your input.csv file inside the input folder. Replace the existing example file.
  • Do not include indexing columns in your file like S.No, Names etc
  • If you have any missing values in your dataset, leave it blank and NEVER mark it 0
  • If you are unsure of the number of clusters you want, keep the program settings at default values.
  • By default, the clustering algorithm chosen is Hierarchical Clustering. Change the method in the program settings file to kmeans for KMeans Clustering.
Running the application

Once you have the two files ready inside the input folder, run the clustering_simulator application

This will generate three files - dendogram.jpg, output.csv and output.txt. Output.csv will have your predicted values and output.txt will give information about the algorithm used, number of clusters and elements per cluster details.

The dendogram.jpg file will help you verify if the number of clusters chosen by the application is reasonable. You can always change the cluster number in the program settings file and run the application again.

Walk-through with an example

Doubts and Clarifications

With every dataset being unique in its own way, your algorithm might require custom configurations and this application may not give you the best results.

And in such cases mail your requirements to khan.mlblog@gmail.com

Download Simulator

Unsupervised ML Simulator v1.0 (Windows only)