0
$\begingroup$

I am working on a churn prediction model, where I am trying to predict probability of employee churn. For each employee I have the following features

1) Role 2) Total experience 3) Current experience etc.

but I don't have labelled historical data about who churned in the past. So supervised learning is not applicable here.

What kind of approaches are useful in these kind of problems.

$\endgroup$
0
$\begingroup$

This problem falls under the category of unsupervised learning, in which the most common approach is clustering:

One tries to divide data in different groups based on the features that are given. Various approaches exist, for example k-means or density-based approaches; take a look at for example sklearn.

Another possiblity is to perform dimensionality reduction in order to visualize your data. This can be done by for example principal component analysis or by constructing a Self-Organizing Map (SOM). The first one is a linear method (which can be extended to a non-linear method, so-called Kernel PCA), the second one is non-linear.

$\endgroup$

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.