# Predicting the Rating of the Next Book Read — solution

*English translation by SOTA – AI Community of the Russian original. Organisers who would like this translation removed can email sota.ai.community@gmail.com.*

## Solution

The solution involves building a regression model that uses the user's interaction history and (optionally) the metadata of books and users.

Possible directions for improvement (these are not requirements):

- building aggregated features per user and per book;
- correct leak-free validation that respects the chronological split;
- extracting features from the text descriptions of books while observing the constraints of the stage.