Improving record matching in imprecise and uncertain datasets

David Croft

Research output: Contribution to journalArticlepeer-review


Museum collections represent a highly challenging search space. This article proposes a novel approach for co-referent record identification which is suitable for use across multiple separate collections. The proposed approach is intended to be suitable for use despite highly imprecise/uncertain attribute values in the records. It is hoped that this can be achieved through a combination of aspects from the fields of probabillistic record linkage, document classification, and fuzzy clustering.
Original languageEnglish
Pages (from-to)347-354
Number of pages8
JournalLiterary and linguistic computing
Issue number4
Publication statusPublished - 2012


Dive into the research topics of 'Improving record matching in imprecise and uncertain datasets'. Together they form a unique fingerprint.

Cite this