Commonsense Knowledge Graph towards Super APP and Its Applications in Alipay.

The recently explosive growth of Super Apps brings great convenience to people's daily life by providing a wide variety of services through mini-programs, including online shopping, travel, finance, and so on. Due to the considerable gap between various scenarios, the restriction of effective information transfer and sharing severely blocks the efficient delivery of online services, potentially affecting the user's app experience. To deeply understand users' needs, we propose SupKG, a commonsense knowledge graph towards Super APP to help comprehensively characterize user behaviors across different business scenarios. In particular, our SupKG is carefully established from multiplex and heterogeneous data source in Alipay (a well-known Super App in China), which also emphasize abundant spatiotemporal relations and intent-related entities to answer the fundamental question in life service "which service do users need at what time and where". On the hand, the successful application of SupKG hinges on the effective form of network representation i.e., Knowledge Graph Embedding (KGE). However, a series of unsatisfying issues still need to be carefully considered in the industrial environment: i) bridging language representations with knowledge structure in a unified manner, ii) alleviating the skewed data distribution in SupKG, and iii) effectively characterizing hierarchical structures in SupKG. With these motivations, we develop a novel knowledge graph representation learning framework for SupKG, enabling various downstream applications to benefit from learned representations of entities and relations. Extensive experiments on the standard knowledge graph completion task demonstrate the consistent and significant performance improvement of our representation learning framework, which also greatly benefits the supplementation of potential knowledge of SupKG. Towards real-world applications in Alipay, our SupKG and learned representations show the potential superiority of integrating global behaviors in cold-start scenarios and providing high-quality knowledge for warming up the graph-based ranking.