AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Ontologue: Declarative Benchmark Construction for Ontological Multi-Label Classification

Sean T. Yang, Bernease Herman, Bill Howe

NeurIPS 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add label text embeddings (BERT) as graph node featuresSurj · Ontologue COMEDYAUPRC: 0.857 -> 0.872Improved
add label text embeddings (BERT) as graph node featuresSurj · Ontologue ENGINEERINGAUPRC: 0.853 -> 0.873Improved
add label text embeddings (BERT) as graph node featuresSurj · Ontologue LAWAUPRC: 0.796 -> 0.802Improved
add label text embeddings (BERT) as graph node featuresSurj · Ontologue MAINAUPRC: 0.809 -> 0.834Improved
add label text embeddingsSurj · Ontologue MAINnormalized GHV: 0.006 -> 0.005Improved
add label text embeddingsSurj · Ontologue ENGINEERINGnormalized GHV: 0.003 -> 0.002Improved
use C-HMCNN instead of SurjSurj · Ontologue ENGINEERINGAUPRC: 0.853 -> 0.842Hurt
use C-HMCNN instead of SurjSurj · Ontologue COMEDYAUPRC: 0.857 -> 0.870Improved
use Surj instead of Naive baseline for hierarchy complianceNaive · Ontologue MAINnormalized GHV: 0.038 -> 0.006Improved
use CLUS-HMC instead of Naive baselineNaive · Ontologue COMEDYAUPRC: 0.839 -> 0.620Hurt
proposed benchmarks vs prior benchmarks in share of trivial or impossible labelsbenchmarks · label difficultyfraction trivial/impossible labels: 75 % -> 45 % to 58 %Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy