The code was developed and compiled via Java
To execute, extract the folder k_means into the directory (NOT the one in source code) and set the following command:
java k_means.clustering
It should output a text file fittingly called 'output.txt' into the directory
Some notes:
-
The program is, by design, formatted to overrite the output.txt file if it already exists. If you require multiple individual files, then move the file out of the directory so that a new one is generated.
-
I was encountering some strange errors while testing later on that was somehow preventing multiple iterations of centroid determination, causing the initial k-groups to remain and causing the groups to display incorrectly. I was unable to figure out why this happened, as it seemed to occur randomly and code changes didn't seem to do anything. As far as I can tell, I fixed it, but I do apologize if it occurs again.
-
I was having trouble with zeus while testing if it worked on that server, so I am only able to say I have majority confidence in this program working properly on that server and not full confidence.