A Pre-trained Data Deduplication Model based on Active Learning

arXiv:2308.00721v2 Announce Type: replace
Abstract: In the era of big data, the issue of data quality has become increasingly prominent. One of the main challenges is the problem of duplicate data, which can arise from repeated entry or the merging of multiple data sources. These "dirty data" problems can significantly limit the effective application of big data. To address the issue of data deduplication, we propose a pre-trained deduplication model based on active learning, which is the first work that …

abstract active learning arxiv become big big data challenges cs.ai cs.lg data data quality data sources deduplication dirty data duplicate issue merging multiple quality type