1
$\begingroup$

Looking at this data-set, I want to illustrate ideally to what degree and in what direction each of the properties (fixed acidity,volatile acidity,citric acid,etc.) influence each other.

Right now I can only think of using a chord diagram or weighted network diagram to illustrate the magnitude of influence properties have on one another.

Are there any reasonable ways to visualise both the magnitude of influence properties have on one another and the direction of influence (positive/negative influence)?

Or any better ways to simply visualise the magnitude of influence?

$\endgroup$
1
$\begingroup$

Using seaborn in python, I usually plot the pairwise distributions with sns.pairplot. seaborn provides a wide variety of data exploration tools in a high level API, it is quite powerful for data exploration.

Here's what sns.pairplot looks like on the popular iris dataset: Pairplot on iris dataset

Sometimes plots like the following (a specific kind of cobweb plot) can be useful too. Cobweb plot

Those represent the distribution of each parameter on each vertical line, with a web of links joining values for a single entry in the dataset.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.