/kmeans-clustering-cpp

C++ implementation of simple k-means clustering algorithm.

Primary LanguageC++

Object Oriented Programming I FALL SEMESTER PROJECT

Abstract

The project includes the implementation of the K-Means that is an unsupervised machine learning algorithm.

plot

Classes

  • Clustering is base class. It designed for utilty function (readfile etc.), keeping vector of points and result of clustering.
  • Kmeans class is source code of algorithm. It designed for iteration over data, maintain K-means++ centroid initialization algorithm.
  • Node class stores data points in vector with format (id,x,y).
  • Visualize class designed for Matplot++ visualization library.

Installation

  • First clone repo in your workspace.

  • For visualizatin process, git clone https://github.com/alandefreitas/matplotplusplus.git

  • sudo apt install gnuplot to see plot after program is terminated.

  • After clone above repository in workspace, run cmake .

  • make

  • run binary file with run ./program

Dependencies

  • cmake version 3.14+ installed
  • make sure your C++ compiler supports C++17:

Tested

  • ubuntu 20.04
  • g++ 9.3.0
  • cmake 3.16.0
  • gnuplot 5.2 patchlevel 8

Demo

  • Demo file consist of result of algorithm with different parameters.