Recurrent knowledge graph embedding for effective recommendation

Knowledge graphs (KGs) have proven to be effective to improve recommendation. Existing methods mainly rely on hand-engineered features from KGs (e.g., meta paths), which requires domain knowledge. This paper presents RKGE, a KG embedding approach that automatically learns semantic representations of both entities and paths between entities...

Effect of Corridor Design on Energy Consumption for School Buildings in the Cold Climate.

This paper discusses the energy impact of corridor design for school buildings in the cold climate of China. Local school buildings were classified into three types in terms of the corridor design patterns. Architectural related parameters of corridors which could have a potential impact on the energy consumption were summarized and discussed,...