I know what you're thinking. That's not k, that's Nemo. This project is not about finding him. I wish it was. But its not. I just needed an excuse to put Nemo's picture.
Moving on...
In this project, we propose, justify and implement (in R) an algorithm to find k for k-means clustering.
- Algorithm.pdf has the detailed explanation of how we find the best k.
- finding.R has the R code of implementation.