How can i plot a database of vectors

I have a database with 3 classes, class 1 =59 vectors, class 2:80, class 3 100. I want to plot those 3 classes of vectors(they have each 13 attributes) with different colors in same chart. Then i want to plot another vector with a different color in the same chart in order to figure out what is the nearest class to this vector. Is that possible?

Here is the database: http://archive.ics.uci.edu/ml/machine-learning-databases/wine/

Answer 1

[Image Analyst]

The vector is a list of wine ID numbers I guess. I think you can pick 3 of the 13 attributes and then plot all the vectors with scatter3(). You can pass in colors and sizes into scatter 3 so pick a different color and size for each of the 3 classes so you can tell each cluster apart once they're plotted on the scatterplot. I don't know how to do more than 3 classes (3 attributes plus the class label) since we can visualize in 3D but not 13D. So you can do 3 attributes at a time. So plot your cluster centroids in one color and size, and then all the indexes from that class in the same color but a smaller size. Repeat for the other two classes.

Here's a video for how to visualize or conceptualize up through the 10th dimension but honestly I get lost after about dimension 4 or 5. https://www.youtube.com/watch?v=JkxieS-6WuA