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

The goal of entity resolution is to identify records in multiple datasetsthat represent the same real-world entity. However, comparing all recordsacross datasets can be computationally intensive, leading to long runtimes. Toreduce these runtimes, entity resolution pipelines are constructed of twoparts: a blocker that applies a computationally cheap method to selectcandidate record pairs, and a matcher that afterwards identifies matching pairsfrom this set using more expensive methods. This paper presents SC-Block, ablocking method that utilizes supervised contrastive learning for positioningrecords in the embedding space, and nearest neighbour search for candidate setbuilding. We benchmark SC-Block against eight state-of-the-art blockingmethods. In order to relate the training time of SC-Block to the reduction ofthe overall runtime of the entity resolution pipeline, we combine SC-Block withfour matching methods into complete pipelines. For measuring the overallruntime, we determine candidate sets with 99.5% pair completeness and pass themto the matcher. The results show that SC-Block is able to create smallercandidate sets and pipelines with SC-Block execute 1.5 to 2 times fastercompared to pipelines with other blockers, without sacrificing F1 score.Blockers are often evaluated using relatively small datasets which might leadto runtime effects resulting from a large vocabulary size being overlooked. Inorder to measure runtimes in a more challenging setting, we introduce a newbenchmark dataset that requires large numbers of product offers to be blocked.On this large-scale benchmark dataset, pipelines utilizing SC-Block and thebest-performing matcher execute 8 times faster than pipelines utilizing anotherblocker with the same matcher reducing the runtime from 2.5 hours to 18minutes, clearly compensating for the 5 minutes required for training SC-Block.

SC-Block: Supervised Contrastive Blocking within Entity Resolution Pipelines