How can I visualise and understand the relationship between n dimensions I am new at the field of machine learning. I have non-linear 6 dimensions, and I want to understand the relationship between 5 dimensions first. And, then understand how these 5 dimensions behave related to the sixth. dimension. Can I ask if there is any techniques in machine learning can help?
 A: You can perform a t-SNE model, which maps N dimensions into two, so you can see how is the inner data structure.
Once you have the t-SNE model, you'll see dots in a 2-dimension matrix, and then you can color by any variable and see how each variable is distributed across the "map".
I wrote a post in R some weeks ago which does this mapping based on a cluster model. The cluster number was my target or dependent variable; you can easily adapt the code available in github for the variable you want:

t-SNE performs very well, here a live example to play done by Google Team: http://distill.pub/2016/misread-tsne
Another beautiful t-SNE visualization, which groups images according to its similarity: https://indico.io/blog/visualizing-with-t-sne:

t-SNE is implemented in many languages (such as Python).
If you want to try another approach (in R), you can analyze the same but based on artificial neural networks, named Self-Organizing Map -SOM-. This model emulates the visual cortex of our brains which maps objects closer if they share similar features. Like saying, "the recognition of houses and dogs are in different parts of the brain". Check this great post made some time ago:

The last image tries to be a piece of brain mapping customers...
As you can see, again you have 2-dimensions, and there is a zone which clearly shows people with high and low education level.
One last approach on this topic, using the well known PCA, principal component analysis: http://setosa.io/ev/principal-component-analysis/
Keep in mind, all of these techniques explores the same "thing" (the data), from different points of view. Some perform better than others depending on the context. Just make you own exploration, good luck!
