Mutual Information-based Preference Disentangling and Transferring for Cross-Domain Recommendations

Watch a 13-minute conference presentation from SIGIR 2024 exploring innovative approaches to non-overlapped multi-target cross-domain recommendations through mutual information-based preference disentangling and transferring techniques. Learn how researchers Zhi Li, Daichi Amagata, Yihong Zhang, Takahiro Hara, Shuichiro Haruta, Kei Yonekawa, and Mori Kurokawa tackle the challenges of cross-domain recommendation systems by developing methods to effectively transfer user preferences across different domains without overlapping items or users.