This repository contains all needed information to execute our experiments:
- Datasets: dbbook and movielens1m, well known in the community.
- Rule miner: we used AMIE - https://github.com/lajus/amie - and we thank the author who provided the community this tool.
- Graph embedding technique: we used a customized version of KALE - https://github.com/iieir-km/KALE. Thanks to the authors who provided the community this source code
- Recommender system architecture: A python notebook (executable with colab) with the recommendation model used in this work, a customized version of AMAR - https://github.com/swapUniba/Deep_CBRS_Amar. Thanks to the authors who provided the community this source code.
To execute experiments and evaluate results, use Elliot: https://github.com/sisinflab/elliot - Thank the authors for this source code.
Please check the readme files in the subfolders for further operative details