Data Preparation, Transforms, Quality, and Management
摘要
Data preparation and feature engineering transform source data elements into a form that can be used by analytic and machine learning methods. Raw source data elements are transformed into data design features that are specified in the data design through an iterative process of mapping data elements to concepts, value sets, and phenotype expressions. Data that meet the data design criteria are extracted into a data mart where the quality of the data can be assessed. Once data are of sufficient quality and meet expectations, ML features are developed for use in machine learning models.