Hands-on Tutorial
Genetic Algorithm based Clustering Algorithm in Searching Robust Initial Centroids for k-means
Optimizing robust initial centroids for k-means using genetic algorithm: a solution for local optima
In the previous article, we learned together about the basic calculations of the Genetic Algorithm in a simple problem case, linear programming. We also understood the importance of initial cluster centroids in the k-means algorithm, as they directly determine the final clusters generated at the end of the process.
Today, we will delve into the application of the Genetic Algorithm in k-means clustering analysis. While the problem of obtaining robust cluster centroids is not crucial nowadays, as sklearn has produced an approach called k-means++ to ensure robustness, it can still arise if we attempt to build the k-means algorithm from scratch without any optimization. This is where the Genetic Algorithm comes into play! The Genetic Algorithm can optimize the k-means algorithm to obtain robust cluster centroids. In this article, we will demonstrate how the Genetic Algorithm can be used to approach k-means++ and the impressive results it can yield.
