Is Plotting a scatter plot for binary class data good or not?
Definitely it is a good idea if one can find out the relationship between features to see if there are any relation between them.
If two features seem related then we can choose anyone instead choosing both features (but I am not sure how it will be related to other dimensions as well) . So, I think it is a good idea to understand relations between different features and see if there is any redundancy in features (correlated). If it doesn’t seem redundant then we can feed it to our mathematical function (also helps in choosing what model can be chosen) to solve binary classification task.
I am working on project and by ploting scatter plot i don’t get any good relation between the features but by using the correlation matrix i get to know that the feature is having a good relation between them.