I have the following problem: I am doing some research on the accuracy of recommender algorithms that are mostly used nowadays.
So, one way to measure their performance is by checking how well they predict a certain value under different sizes of a given dataset, meaning, sparsity in a ratings matrix.
I need to find a way to calculate the root mean square error(or mae), some metric, versus the sparsity in the dataset. As an example, let’s have a look at the picture below:
You can see that it says:
“RMSE as a function of sparsity. 5000 ratings were removed from the training set(initially containing 80000 ratings) in every iteration. “
I’m using Python and the Movielens dataset. Do you know how can I achieve this in the mentioned language? Is there any tool to do that?