… the above challenges, we propose a Meta-Graph based Attention-aware Recommendation
(MGAR) … method which utilizes rich meta-graph based latent features to make heterogeneous …

… to integrate meta-graph embeddings, which … based recommender system, called attention
aggregator. The weight obtained for a meta-graph represents the importance of the meta-graph…

… networking embedding based methods, this paper proposes a meta-graph embedding method
for item recommendation … Specifically, we first extend the meta-path to the meta-graph to …

… recommendation algorithm, a meta-based fusion heterogeneous information network
recommendation … The algorithm integrates the characteristics of multi-relationship social network …

… Specifically, we run experiments to compare RMSE of each metagraph in Figures 4 and 5.
The RMSE of each metagraph is shown in Figure 9. Note that we show for comparison the …

… 45] have been proposed to employ heterogeneous information network to model the interactive
local relations to address early summarization problem. With the help of meta-path, Jin et …

… method to predict ratings, which can flexibly integrate heterogeneous information through …
observed ratings to effectively select useful meta-graph based features for rating prediction. …

… In this paper, we proposed a recommendation method over Heterogeneous Information
Networks (HINs), named MGBE. MGBE learns the semantic representations of meta-graphs and …

… We conduct extensive experiments on three large real-world heterogeneous information
networks to verify the advantages of our algorithm. As indicated by experimental results, our …

… model over heterogeneous information network for personalized recommendation. In: …
Zhao H, Yao Q, Li J, Song Y, Lee D (2017) Meta-graph based recommendation fusion over …