I am currently doing an analysis in which I need to "profile" each record. For example, let's say I have a dataset of accounts with customer information (name, id, address, money spent, products bought, etc).In my real case there are 100+ features. The customers are already labelled as "good" or "bad". A good customer is whomever spent more than $10k for example. However, before jumping into building a classification model, I'd like to profile each customer to answer the questions: How does a "good" customer look like? How does a "bad" customer look like?. What would be the analysis techniques that could be used for this scenario?
3
-
$\begingroup$ how about clustering techniques on the dataset? $\endgroup$– Nikos M.Commented Jun 29, 2022 at 15:00
-
$\begingroup$ An efficient solution is to do random forest classification in order to group the main scenarios automatically. machinelearningmastery.com/random-forest-ensemble-in-python $\endgroup$– Nicolas MartinCommented Jun 29, 2022 at 19:57
-
$\begingroup$ So doing unsupervised learning on the dataset seems the way to go $\endgroup$– thesadclownCommented Jul 1, 2022 at 14:19
Add a comment
|