dainiusjocas / data-mining Goto Github PK

require 'file.rb' doesn't work

Do we need to compute average value only of numeric attributes or we also should put into vector most frequent nominal value within cluster?

When using 1.9.2 version of ruby, require 'file.rb' function doesn't work.

If you want to see your name in commit history you should write the commands in terminal:
git config --global user.name "Your Name"
git config --global user.email [email protected]

DBSCAN didn't cluster the full adult dataset in >2h.
A smaller dataset causes K-means to cluster everything into one cluster...

Minkowski distance should be implemented, so we could evaluate and compare the algorithms more precisely. By default the Manhattan distance should be used (p=1), but also a user should be able to choose euclidian distance (p=2) or any other value of parameter p.

This issue is just for presentation.