Readme.md

In cogtepsum/DDP_Assignment: K-Means Example

This is a ShinyApp created for the course project of Developing Data Products course from Cousera Data Science Specialization.

Source code is available on GitHub: https://github.com/cogtepsum/DDP_Assignment

This is a very simple app that may help one to understand K-Means clustering algorithm. It generates some random data, distributed normally across centers, number of which is determined by the user as a number of "true clusters". Positions of centers are generated from uniform random distribution. Variance of all of these random variables is adjusted according to the number of true clusters.

Instructions:

So, basically, you choose a number of Gaussian distribution to generate, and app will create a combined distribution of them. Afterwards you choose what is the number of clusters you want this data to be divided into - and get your clustering plot.

Citation:

Winston Chang, Joe Cheng, JJ Allaire, Yihui Xie and Jonathan McPherson (2015). shiny: Web Application Framework for R. R package version 0.12.2. https://CRAN.R-project.org/package=shiny

Maechler, M., Rousseeuw, P., Struyf, A., Hubert, M., Hornik, K. (2015). cluster: Cluster Analysis Basics and Extensions. R package version 2.0.3.

H. Wickham. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, 2009.

cogtepsum/DDP_Assignment documentation built on May 13, 2019, 8:49 p.m.