Most existing recommendation methods assume that all the items are provided by separate producers
rather than users. However, it could be inappropriate in some recommendation tasks since users may
generate some items. Considering the userโitem generation relation may benefit recommender systems
that only use implicit userโitem interactions. However, it may suffer from a dramatic imbalance.
The number of userโitem generation relations may be far smaller than the number of userโitem
interactions because each item is generated by at most one user. At the same time, this item can be
interacted with by many users. To overcome the challenging imbalance issue, we propose a novel Deep
Interaction-Attribute-Generation (DIAG) model. It integrates the userโitem interaction relation, the
userโitem generation relation, and the item attribute information into one deep learning framework.
The novelty lies in the design of a new itemโitem co-generation network for modeling the userโitem
generation information. Then, graph attention network is adopted to learn the item feature vectors
from the userโitem generations and the item attribute information by considering the adaptive impact
of one item on its co-generated items. Extensive experiments conducted on two real-world datasets
confirm the superiority of the DIAG method.