AI

An Explainable Header-Centric Framework for Large-Scale Semantic Table Interpretation and Data Quality Assessment

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 09/10/2026, 04:03:23 UTC.

Briefings

This paper introduces an explainable header-centric framework for semantic table interpretation in metadata-only scenarios. The system maps column headers to 39 interpretable FinalFormat types using curated lexical resources, enabling data quality assessment through Data Quality Issues (DQIs) like missing data or type mismatches. Evaluated across 120,000 columns from diverse benchmarks including UCI, Kaggle, and SemTab 2024, the framework provides practical coverage for noisy real-world metadata while supporting KG alignment to DBpedia and Schema.org. A blinded diagnostic audit on the SemTab 2024 track suggests many reported mismatches stem from benchmark granularity and ontology effects rather than header-centric prediction failures.

Key points

• The framework assigns 39 interpretable FinalFormat types to column headers using curated lexical resources for semantic annotation.
• Data Quality Issues (DQIs) such as missing data, duplicates, and type violations are detected via header-centric validation rules.

Context and limitations

Evaluation was limited to 120,000 columns across heterogeneous benchmarks; results are not revised benchmark performance but diagnostic evidence from a blinded audit.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 09/10/2026, 04:37:35 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeifzj3mesdolw26ljvoo3qdmggwdn7xt6aklxzqvjoevtddoeig72e