CCHZ-DISO: A Timely New Assessment System for Data Quality or Model Performance From Da Dao Zhi Jian

Abstract

With the rapid development of big data, assessment of data quality or model performance has become a hot scientific question. However, most existing lots of metrics focus on specific aspects of the assessment, and comprehensive assessment is rare. Therefore, it is very necessary to develop new assessment system. To address this problem, a new assessment system is constructed which is named after Chen, Chen, Hu, and Zhou (CCHZ)-distance between indices of simulation and observation (DISO) according to the contributions of Xi Chen, Deliang Chen, Zengyun Hu, and Qiming Zhou. CCHZ-DISO system builds on the Euclidean Distance and flexible determination of statistical metrics and their numbers. Due to its simplicity and flexibility, CCHZ-DISO can be readily and widely applied to any subject of science. Therefore, it follows the principle of the Chinese philosopher Lao Zi's Da Dao Zhi Jian which means that the most basic truth is very simple.