Database Dereplication

Database Dereplication is a data curation process that identifies and removes redundant or near-identical entries within a reference database to ensure uniqueness and accuracy of records.

4

Publications

246

Citations

11

Authors

1

Institutions

Publications per year

2017–2026

0

Authors

11

  1. KL
    Keith L. Monson

    Federal Bureau of Investigation, United States

    1 publications · 116 citations

    1

    h-index

  2. JA
    Joseph A. DiZinno

    Federal Bureau of Investigation, United States

    1 publications · 116 citations

    1

    h-index

  3. DG
    David G. Corley

    University of Missouri, United States

    1 publications · 77 citations

    1

    h-index

  4. KM
    Kevin Miller

    Federal Bureau of Investigation, United States

    1 publications · 116 citations

    1

    h-index

  5. DU
    Domenico Ursino

    University of Reggio Calabria, Italy

    1 publications · 19 citations

    1

    h-index

1–5 of 11

Institutions

1

  1. University of Calabria

    Rende, Italy

    1 h-index

    3 publications

    57 citations

1–1 of 1

Venues

  1. Forensic Science Communications

    1 publications · 116 citations

    1

    h-index

  2. Data & Knowledge Engineering

    DKE · 1 publications · 19 citations

    1

    h-index

  3. Journal of Natural Products

    1 publications · 77 citations

    1

    h-index

  4. Science

    1 publications · 34 citations

    1

    h-index

1–4