Abstract
A new framework to formulate and quantify the epistasis of a problem is proposed. It is based on Shannon's information theory. With the framework, we suggest three epistasis-related measures: gene significance, gene epistasis, and problem epistasis. The measures are believed to be helpful to investigate both the individual epistasis of a gene group and the overall epistasis that a problem has. The experimental results on various well-known problems support it. © Springer-Verlag Berlin Heidelberg 2003.
Cite
CITATION STYLE
Seo, D. I., Kim, Y. H., & Moon, B. R. (2003). New entropy-based measures of gene significance and epistasis. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2724, 1345–1356. https://doi.org/10.1007/3-540-45110-2_9
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
