How to use User features(such as age, demographics) and Item features while Creating the Sequential Recommendation System using BERT4Rec?

[np-n]

I want to train a next-procedure prediction for the medical dataset. After checking the documentation and testing the data loading pipeline, what I have figured out is: the emphasis is done on the sequence of item interaction for users along with the timestamp in the loaded data. But there is no consideration of user features and item features.

But, what I know is user features and item features also play a role, In my case, user features are very critical. How I can use such features to design the sequential recommendation system? Thank you!

[Sherry-XLL]

Hello @np-n, thanks for your attention to RecBole!

For sequential recommendation, you can refer to the implementations of GRU4RecF and SASRecF for more details about adding item attributes to recommendation models. Similarly, you can use user features by feature embedding like item features.

Feel free to contact us if you have further questions.

[thigazholi]

Can you please explain further as to how to use user features similar to how we construct item embeddings using item features in SASRecF?