Work

Best practices for conducting benchmarking in the most comprehensive and reproducible way

Year: 2017

Type: preprint

Abstract: Computational biology is rapidly advancing thanks to the many new tools developed and published each month. A systematic benchmarking practice would help biomedical researchers leverage this technological expansion to optimize their projects. Several aspects of algorithm publication and distribution contribute to this challenge. We address these challenges and present seven principles to guide researchers in designing a benchmarking study. Our proposed steps show how benchmarking can create a framework for comparison of newly published algorithms.

DOI: 10.7287/peerj.preprints.3236

Authors: Serghei Mangul, Lana S. Martin, Margaret Distler, Eleazar Eskin, Jonathan Flint

Institution: University of California, Los Angeles

Language: English

Topic: Scientific Computing and Data Management

Subfield: Information Systems and Management

Field: Decision Sciences

Domain: Social Sciences

Open Access status: gold

OpenAlex ID: W4256594036

Cites:

Cited by:

Related to: 10