Seed Set Evaluation

Seed Set Evaluation is a systematic assessment procedure used in hybrid performance testing to determine the representativeness and quality of an initial subset of test instances—selected to seed automated or iterative testing processes—by evaluating how well they capture the underlying characteristics of the broader test population.

198

Publications

8.1K

Citations

551

Authors

237

Institutions

Publications per year

2017–2026

28

Authors

551

  1. DR
    Douglas R. Dewey

    Utah State University, United States

    3 publications · 53 citations

    3

    h-index

  2. GW
    Glenn W. Burton

    University of Georgia, United States

    3 publications · 38 citations

    3

    h-index

  3. LW
    Lonnie W. Aarssen

    Queen's University, Canada

    3 publications · 97 citations

    3

    h-index

  4. RJ
    R. J. Lambert

    University of Illinois Urbana-Champaign, United States

    2 publications · 259 citations

    2

    h-index

  5. MB
    Michael Bienek

    Advanced Micro Devices (Canada), Canada

    2 publications · 44 citations

    2

    h-index

1–5 of 551

Venues

  1. Crop Science

    21 publications · 660 citations

    14

    h-index

  2. American Journal of Botany

    15 publications · 674 citations

    14

    h-index

  3. Annals of Botany

    7 publications · 274 citations

    7

    h-index

  4. Canadian Journal of Plant Science

    CJPS · 5 publications · 128 citations

    5

    h-index

  5. International Journal of Plant Sciences

    IJPS · 5 publications · 110 citations

    5

    h-index

1–5