Presents a supervised record linkage system with BM25 blocking and metadata/citation features to clean noisy scholarly metadata, outperforming title-similarity baselines on CiteSeerX matching tasks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DL 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Cleaning Noisy and Heterogeneous Metadata for Record Linking Across Scholarly Big Datasets
Presents a supervised record linkage system with BM25 blocking and metadata/citation features to clean noisy scholarly metadata, outperforming title-similarity baselines on CiteSeerX matching tasks.