English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Graph Convolution Network (GCN) has attracted significant attention andbecome the most popular method for learning graph representations. In recentyears, many efforts have been focused on integrating GCN into the recommendertasks and have made remarkable progress. At its core is to explicitly capturehigh-order connectivities between the nodes in user-item bipartite graph.However, we theoretically and empirically find an inherent drawback existed inthese GCN-based recommendation methods, where GCN is directly applied toaggregate neighboring nodes will introduce noise and information redundancy.Consequently, the these models' capability of capturing high-orderconnectivities among different nodes is limited, leading to suboptimalperformance of the recommender tasks. The main reason is that the the nonlinearnetwork layer inside GCN structure is not suitable for extracting non-sematicfeatures(such as one-hot ID feature) in the collaborative filtering scenarios.In this work, we develop a new GCN-based Collaborative Filtering model, namedRefined Graph convolution Collaborative Filtering(RGCF), where the constructionof the embeddings of users (items) are delicately redesigned from severalaspects during the aggregation on the graph. Compared to the state-of-the-artGCN-based recommendation, RGCF is more capable for capturing the implicithigh-order connectivities inside the graph and the resultant vectorrepresentations are more expressive. We conduct extensive experiments on threepublic million-size datasets, demonstrating that our RGCF significantlyoutperforms state-of-the-art models. We release our code athttps://github.com/hfutmars/RGCF.

RGCF: Refined Graph Convolution Collaborative Filtering with concise and expressive embedding