Contingency Space

Home How To Metrics About

Citation: A. Ahmadzadeh, D. J. Kempton, P. C. Martens, R. A. Angryk, (under review) "Contingency Space: A Semimetric Space for Classification Verification".



What is Contingency Space? Contingency Space is a performance verification framework introduced recently to help analysis of model performance by providing a multidimensional view for the outcome of the single-value metrics such as accuracy, f1 score, etc. Contingency Space is a bounded semimetric space that provides a generic representation for any performance evaluation metric by visualizing its bivariate distribution function as a surface. Using this online prototype we are trying to make an inventory of the needs in using this framework.

For suggestions, please contact us at aahmadzadeh1[at]gsu[dot]edu.

Developer



Egill Ragnar Gunnarsson


M.S. Graduate from Georgia State University

B.S., Computer Science, Georgia State University, 2020

M.S., Data Science and Analytics — Concentration in Big Data and Machine Learning, Georgia State University