Dual-Target Cross-Domain Bundle Recommendation

The conventional Cross-Domain Recommendation(CDR) approaches are single-target that focus only on improving the recommendation performance of the target domain. To enhance the performance of both source and target domains, dual-target CDR approaches have been proposed. However, existing approaches considered recommending only a single item to users. Besides, they cannot effectively combine the features on both domains. To this end, we propose a novel bundle graphical and attentional model named Dual-Target Cross-Domain Bundle Recommendation(DT-CDBR). Specifically, we first integrate user-item, user-bundle interaction, and bundle-item affiliation into a heterogeneous graph. Then, we assign different weights for both domains via an attention mechanism, and combine the features of common users based on the weights. In this way, our DT-CDBR can dynamically adjust the weights of two domains and further enrich the representation of features. Extensive experiments conducted on real-world datasets demonstrate that our DT-CDBR can improve the recommendation performance on both domains simultaneously.