Multi-View Intent Disentangle Graph Networks for Bundle Recommendation

Bundle recommendation aims to recommend the user a bundle of items as whole. Previous models capture user’s preferences on both and association items. Nevertheless, they usually neglect diversity intents adopting fail disentangle in representations. In real scenario recommendation, intent may be naturally distributed different bundles that (Global view). And contain multiple (Local Each view has its advantages for disentangling: 1) global view, more are involved present each intent, which can demonstrate preference under clearly. 2) The local reveal between since within same highly correlated other. To this end, paper we propose novel model named Multi-view Intent Disentangle Graph Networks (MIDGN), is capable precisely comprehensively capturing items’ associations at finer granularity. Specifically, MIDGN disentangles from two perspectives, respectively: taking Global coupled with inter-bundle items; Local bundle. Meanwhile, compare disentangled views by contrast method improve learned intents. Extensive experiments conducted benchmark datasets outperforms state-of-the-art methods over 10.7% 26.8%, respectively.