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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Generating the Top-N recommendations from a large corpus is computationallyexpensive to perform at scale. Candidate generation and re-ranking basedapproaches are often adopted in industrial settings to alleviate efficiencyproblems. However it remains to be fully studied how well such schemesapproximate complete rankings (or how many candidates are required to achieve agood approximation), or to develop systematic approaches to generatehigh-quality candidates efficiently. In this paper, we seek to investigatethese questions via proposing a candidate generation and re-ranking basedframework (CIGAR), which first learns a preference-preserving binary embeddingfor building a hash table to retrieve candidates, and then learns to re-rankthe candidates using real-valued ranking models with a candidate-orientedobjective. We perform a comprehensive study on several large-scale real-worlddatasets consisting of millions of users/items and hundreds of millions ofinteractions. Our results show that CIGAR significantly boosts the Top-Naccuracy against state-of-the-art recommendation models, while reducing thequery time by orders of magnitude. We hope that this work could draw moreattention to the candidate generation problem in recommender systems.

Candidate Generation with Binary Codes for Large-Scale Top-N Recommendation