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...

Electrochemistry of zirconium in molten chlorides

In this work, the electrochemical behavior of zirconium was studied on an inert molybdenum electrode at 550 °C in a LiCl-KCl-K₂ZrF₆ molten salt system, which is considered as an ideal electrolyte for the zirconium electrorefining process. Several transient electrochemical techniques were used such as cyclic voltammetry,...

MRLR: Multi-level representation learning for personalized ranking in recommendation

Representation learning (RL) has recently proven to be effective in capturing local item relationships by modeling item co-occurrence in individual user's interaction record. However, the value of RL for recommendation has not reached the full potential due to two major drawbacks: 1) recommendation is modeled as a rating prediction problem...

Erosion potential of the Yangtze Delta under sediment starvation and climate change

Deltas are widely threatened by sediment starvation and climate change. Erosion potential is an important indicator of delta vulnerability. Here, we investigate the erosion potential of the Yangtze Delta. We found that over the past half century the Yangtze's sediment discharge has decreased by 80% due to the construction of >50,000 dams...