Adapting User Preference to Online Feedback in Multi-round Conversational Recommendation

Description

This repository is the source code of the paper Adapting User Preference to Online Feedback in Multi-round Conversational Recommendation implemented via PyTorch.

Code

The codes of Yelp (enumerated questions) and LastFM (binary questions) are stored in different branch with minor differences.

Requirement

torch==1.4.0
torch_geometric==1.4.3
tqdm
sklearn

Dataset

The dataset download link: https://drive.google.com/file/d/1qUsdTGHPqawgJ04wx0YtfF8GCRRQYDav/view?usp=sharing The dataset we used is based on Yelp and LastFM which are processed in EAR WSDM'20

After downloading Data.zip, unzip it and put the files in Data/Yelp_data/ or Data/LastFM_data/ to corresponding /data/ folder.

Acknowledgement

Thanks the authors of EAR WSDM'20 for sharing their codes, datasets and model parameters.

Here is the link to their work.