Context: it is necessary to personalize content and offers, and for this purpose it is possible to use information about the user's interaction with the system.
Decision: a hierarchical temporal model of soft client clustering is constructed.
The pilot project was conducted and integrated into the customer's business process:
Identified topics 80% of bank card users;
85% of consumption profiles are interpreted;
Predicting socio-demographic parameters of the audience;
A two-level hierarchy of customer consumption profiles is constructed;
Solved the problem of searching for similar consumers "Look-a-like".
To build the model, we used:
Bank customer transactions;
Description of MSS codes and merchants;
Accompanying information about the user;
Results of sales of banking products.
Model the behavior of users of the Bank;
Probability prediction model before purchasing a banking product;
Segments of the client base.
Customer: Finance, Banking
Technology stack: TopicNet, BigARTM, gensim, Python